Bydureon- INN -exenatide · FPG Fasting plasma glucose GI Gastrointestinal GLP-1 Glucagon-like...

30 Churchill Place ● Canary Wharf ● London E14 5EU ● United Kingdom

An agency of the European Union

Telephone +44 (0)20 3660 6000 Facsimile +44 (0)20 3660 5520

Send a question via our website www.ema.europa.eu/contact

20 July 2017 EMA/538242/2017

Committee for Medicinal Products for Human Use (CHMP)

Assessment report

Bydureon

International non-proprietary name: exenatide

Procedure No. EMEA/H/C/002020/0041

Note

Assessment report as adopted by the CHMP with all information of a commercially confidential nature

deleted.

Assessment report

EMA/538242/2017 /51

Table of contents

1. Background information on the procedure .............................................. 4

1.1. Type II variation .................................................................................................. 4

1.2. Steps taken for the assessment of the product ........................................................ 4

2. Scientific discussion ................................................................................ 5

2.1. Introduction ........................................................................................................ 5

2.2. Non-clinical aspects .............................................................................................. 6

2.2.1. Ecotoxicity/environmental risk assessment ........................................................... 6

2.3. Clinical aspects .................................................................................................... 6

2.3.1. Introduction...................................................................................................... 6

2.3.2. Main study ....................................................................................................... 7

2.3.3. Discussion on clinical efficacy ............................................................................ 27

2.3.4. Conclusions on the clinical efficacy .................................................................... 30

2.4. Clinical safety .................................................................................................... 31

2.4.1. Discussion on clinical safety .............................................................................. 37

2.4.2. Conclusions on clinical safety ............................................................................ 39

2.4.3. PSUR cycle ..................................................................................................... 39

2.5. Risk management plan ....................................................................................... 39

2.6. Update of the Product information ........................................................................ 45

2.6.1. User consultation ............................................................................................ 45

3. Benefit-Risk Balance ............................................................................. 46

4. Recommendations ................................................................................. 50

5. EPAR changes ....................................................................................... 51

Assessment report

EMA/538242/2017 /51

List of abbreviations

Abbreviation or

special term

Explanation

ALT Alanine aminotransferase

BMI Body mass index

BP Blood pressure

CrCl Creatinine clearance

CV Cardiovascular

DBP Diastolic blood pressure

eGFR Estimated creatinine-based glomerular filtration rate

EQW Exenatide once weekly

FPG Fasting plasma glucose

GI Gastrointestinal

GLP-1 Glucagon-like peptide-1

GLP-1RA Glucagon-like peptide-1 receptor agonist

HbA1c Haemoglobin A1c

HDL High-density lipoprotein

LS Least-squares

MTT Meal Tolerance Test

MMRM Mixed model for the repeated measures

od Once daily

PPG Postprandial glucose

SBP Systolic blood pressure

SDT Single dose tray

SGLT2 Sodium glucose cotransporter 2

SU Sulphonylurea

T2DM Type 2 diabetes mellitus

TZD Thiazolidinedione

ULN Upper limit of normal

UTI Urinary tract infection

Assessment report

EMA/538242/2017 /51

1. Background information on the procedure

1.1. Type II variation

Pursuant to Article 16 of Commission Regulation (EC) No 1234/2008, AstraZeneca AB submitted to the

European Medicines Agency on 5 December 2016 an application for a variation.

The following variation was requested:

Variation requested Type Annexes

affected

C.I.6.a C.I.6.a - Change(s) to therapeutic indication(s) - Addition

of a new therapeutic indication or modification of an

approved one

Type II I and IIIB

Extension of indication for Bydureon to include the add-on use of exenatide in combination with

dapagliflozin to patients whose diabetes is not adequately controlled with metformin based on the study

D5553C00003 (Duration 8 study); section 4.1 of the SmPC is updated in order to align the indication

wording with more recently approved glucose-lowering agents. Section 5.1 of the SmPC is also updated

with the results of study D5553C00003 (Duration 8 study). The Package Leaflet is updated accordingly.

In addition, the Marketing authorisation holder (MAH) took the opportunity to make minor editorial

changes in the SmPC and Package Leaflet. Furthermore, the updated RMP version 24 has been submitted.

The variation proposed amendments to the Summary of Product Characteristics and Package Leaflet.

Information on paediatric requirements

Pursuant to Article 8 of Regulation (EC) No 1901/2006, the application included an EMA Decision

P/0130/2016 on the agreement of a paediatric investigation plan (PIP).

At the time of submission of the application, the PIP (P/0130/2016) was not yet completed as some

measures were deferred.

Information relating to orphan market exclusivity

Similarity

Pursuant to Article 8 of Regulation (EC) No. 141/2000 and Article 3 of Commission Regulation (EC) No

847/2000, the applicant did not submit a critical report addressing the possible similarity with authorised

orphan medicinal products because there is no authorised orphan medicinal product for a condition

related to the proposed indication.

Scientific advice

The MAH did not seek Scientific Advice at the CHMP.

1.2. Steps taken for the assessment of the product

The Rapporteur and Co-Rapporteur appointed by the CHMP were:

Assessment report

EMA/538242/2017 /51

Rapporteur: Kristina Dunder Co-Rapporteur: N/A

Timetable Actual dates

Submission date 5 December 2016

Start of procedure: 24 December 2016

CHMP Rapporteur Assessment Report 17 February 2017

PRAC Rapporteur Assessment Report 17 February 2017

PRAC members comments 01 March 2017

PRAC Outcome 9 March 2017

CHMP members comments 13 March 2017

Updated CHMP Rapporteur(s) (Joint) Assessment Report 16 March 2017

Request for supplementary information (RSI) 23 March 2017

CHMP Rapporteur Assessment Report 26 June 2017

CHMP members comments 10 July 2017

Updated CHMP Rapporteur Assessment Report 12 July 2017

Opinion 20 July 2017

2. Scientific discussion

2.1. Introduction

The purpose of this application is to provide information on the efficacy and safety of concomitant add-on

treatment with exenatide and dapagliflozin as add-on to metformin, supporting addition of new study

data to the Summary of Product Characteristics (SmPC), section 5.1.

As the combination with SGLT2 inhibitors is not covered by the current wording of the indication, an

update of SmPC section 4.1 has been proposed, and the MAH proposed at the same time to simplify the

wording to be in line with more recently approved glucose-lowering agents including other glucagon-like

peptide-1 receptor agonists (GLP-1RAs).

Exenatide once weekly

Exenatide, a GLP-1RA, exerts its glycaemic-lowering effect by enhancing glucose-dependent insulin

secretion, suppressing glucagon release, and delaying gastric emptying. In addition, it lowers weight by

inducing satiety through central mechanisms.

Dapagliflozin

Dapagliflozin is a selective inhibitor of sodium glucose cotransporter 2 (SGLT2). SGLT2 inhibition leads to

pharmacologically controlled glucosuria, which leads to subsequent glycaemic effects, including lowering

of FPG and PPG, and reductions in HbA1c and weight. The glucosuria results in a mild osmotic diuresis

which is associated with reductions in SBP.

Rationale for combination therapy with exenatide and dapagliflozin

Assessment report

EMA/538242/2017 /51

Type 2 diabetes is a progressive disease that is characterised by defects in multiple organ systems and

usually requires combination therapy with agents that target different pathways. Because physiological

defects may be manifested differently in individual patients, and patients differ in their care needs due to

differences in demographics, comorbidities, individual preferences and other factors, individualisation of

diabetes management is emphasised in international guidelines as essential for successful diabetes care.

The availability of various types and combinations of diabetes therapy is consistent with this emphasis on

patient-centered care.

The combination of exenatide and dapagliflozin is expected to have greater glucose-, weight- and blood

pressure (BP)-lowering effects compared to the individual agents due to complimentary mechanisms of

action (MOAs). Dapagliflozin reduces plasma glucose in an insulin-independent fashion by causing

excretion of urinary glucose, in contrast to the insulin-dependent mechanisms of glucose lowering with

exenatide, including stimulation of insulin release and glucagon suppression. The urinary loss of calories

that leads to weight loss with dapagliflozin may be associated with a compensatory increase in food

intake; this effect may be counteracted by the increase in satiety that occurs with exenatide. In addition,

the BP-lowering effects of dapagliflozin and exenatide are also likely due to at least partially differing

mechanisms: dapagliflozin leads to osmotic diuresis and natriuresis, whereas the mechanism by which

exenatide lowers SBP is unknown, potentially due to vasodilation and/or natriuresis. Both agents act via

glucose-dependent mechanisms and do not intrinsically increase the risk of hypoglycaemia.

The clinical programme for this submission consists of a single Phase 3 study, Study D5553C00003. The

efficacy and safety data from the 28-week treatment period of this study support concomitant add-on

treatment with the combination of EQW 2 mg + dapagliflozin 10 mg od in patients with T2DM who have

inadequate glycaemic control on metformin.

2.2. Non-clinical aspects

No new clinical data have been submitted in this application, which was considered acceptable by the

CHMP.

2.2.1. Ecotoxicity/environmental risk assessment

No environmental risk assessment was performed because exenatide as a moderately sized naturally-

occuring peptide is unlikely to result in significant risk to the environment, in line with the current ERA

guideline (CPMP/SWP/4447/00).

2.3. Clinical aspects

2.3.1. Introduction

GCP

The Clinical trials were performed in accordance with GCP as claimed by the applicant.

The applicant has provided a statement to the effect that clinical trials conducted outside the community

were carried out in accordance with the ethical standards of Directive 2001/20/EC.

• Tabular overview of clinical studies

Study ID Objectives

of the

Design and

duration

Study drugs

Background

Number of

subjects

Gender

(M/F)

Population

Assessment report

EMA/538242/2017 /51

study therapy

Route of

administration

randomized/

treated

Mean age

(range)

D5553C00003

(DU

RATIO

N 8

) Efficacy

and safety

Randomized,

double-blind,

active-controlled,

multi-center,

Phase 3

Duration:

28 weeks

(finalised), with

24- and 52- week

extensions

(ongoing)

EQW 2 mg

(injection) + Dapa

10 mg (oral)

versus

EQW 2 mg

(injection) + Dapa

Placebo (oral)

versus

Dapa 10 mg (oral)

+ EQW Placebo

(injection)

Background:

Met ≥1500mg/day

EQW/Dapa:

231/231

EQW/Placebo:

231/230

Dapa/Placebo:

233/233

48% M

52% F

54 yrs (26

to 80 yrs)

T2DM, ≥18

years,

HbA1c ≥8%

to ≤12% on

Met, CrCl

≥60 mL/min

2.3.2. Main study

A 28-week, Multicentre, Randomised, Double-Blind, Active-Controlled, Phase 3 Study with a

24-week Extension Phase Followed by a 52-week Extension Phase to Evaluate the Efficacy and

Safety of Simultaneous Administration of Exenatide Once Weekly 2 mg and Dapagliflozin Once

Daily 10 mg Compared to Exenatide Once Weekly 2 mg Alone and Dapagliflozin Once Daily 10

mg Alone in Patients with Type 2 Diabetes who have Inadequate Glycaemic Control on

Metformin

Methods

Study D5553C00003 was a 28-week, randomized, double-blind, active-controlled, multicentre, Phase III

efficacy and safety study with a 24- and subsequent 52-week extension of simultaneous administration of

EQW 2 mg and dapagliflozin 10 mg QD compared to EQW 2 mg alone and dapagliflozin 10 mg QD alone

in T2DM patients with inadequate glycaemic control on metformin.

Assessment report

EMA/538242/2017 /51

Figure 1 Flow chart of study design

EQW Exenatide once weekly; HbA1c Haemoglobin A1c

The study consisted of a Screening Visit (Visit 1), a 1-week placebo Lead-in period, a Randomization Visit

(Visit 3), and 9 further visits at 1- to 4-week intervals during a treatment period of 28 weeks. At the end

of treatment, patients were to enter a 24-week extension period (Extension Period 1), with 3 further

visits at 8-week intervals and a subsequent 52-week extension period (Extension Period 2) with 4 further

visits at 13-week intervals. A follow-up visit was conducted 10 weeks after the last dose of study

medication in Extension Period 2.

Study participants

The population selected for this study were patients with T2DM and inadequate glycaemic control on

metformin, with a stable dose of metformin ≥1500 mg/day for at least 2 months prior to screening. The

HbA1c inclusion criterion at randomisation (ie, 8.0% to 12.0%, inclusive) was selected to include patients

with poor glycaemic control.

The exclusion criterion at randomisation for creatinine clearance was <60 mL/min (1 mL/s) (calculated by

Cockcroft-Gault formula) or a measured serum creatinine value of ≥133 µmol/L for male patients and

≥124 µmol/L for female patients. These exclusion criteria are consistent with prescribing guidelines for

metformin, dapagliflozin, and EQW.

The purpose of the majority of the inclusion and exclusion criteria was to limit confounding factors that

may complicate the interpretation of the study results (e.g., corticosteroid-induced T2DM,

haemoglobinopathies that would interfere with the HbA1c analyses), or to exclude patients whose safety

could be compromised by participation in the study.

Treatments

This was a double-blind, active-controlled study. The comparison of the combination of dapagliflozin and

exenatide with its individual components is consistent with regulatory guidance regarding the

investigation of combination products (CHMP 2009).

Assessment report

EMA/538242/2017 /51

Background therapy

Metformin is recommended as the initial pharmacological therapy in both the United States (US) and the

European Union (EU).

Dapagliflozin

The 10 mg dose was chosen for this study as it has been extensively studied in Phase III trials and has

demonstrated a favourable benefit-risk profile. In addition, it is the most effective approved dose and

therefore the most appropriate option for patients with very poor glycaemic control.

Exenatide

The 2 mg dose was used for this study as it is the dose that was studied in the Phase III program, and is

the only approved dose of exenatide once weekly. EQW was provided as a kit containing 1 vial of

exenatide powder for injection, 1 syringe with diluents, 2 vial adaptors, and 2 23-Gauge x 5/16 inch

needles. i.e. as SDT.

Rescue

During 28-Week randomised treatment period, patients with inadequate glycaemic control based on

progressively stricter glycaemic criteria (Table 1) remained in the study and received open-label rescue

therapy with basal insulin while they continued receiving study medication.

Table 1 Criteria for initiation of rescue therapy during the randomised 28-week

treatment period

Patients who meet the rescue criteria during the 28-Week treatment period had to complete Rescue visit

procedures (equivalent to the Week 28 assessments) before receiving open-label rescue therapy.

Rescued patients with central laboratory HbA1c values >8.0% despite a maximum tolerated dose of

rescue therapy for 12 weeks were to be discontinued from the study and referred for additional anti-

hyperglycaemic therapy.

Objectives

Primary objective

To compare the change from baseline in HbA1c at 28 weeks between exenatide once weekly (EQW) 2 mg

and dapagliflozin 10 mg administered simultaneously compared to EQW 2 mg alone and dapagliflozin 10

mg alone.

Secondary objectives

To compare the effect of EQW+dapagliflozin to EQW+placebo and/or to dapagliflozin+placebo, on

changes in glycaemic control and anthropometric measures.

Assessment report

EMA/538242/2017 /51

Outcomes/endpoints

Primary endpoint

The primary endpoint was change in HbA1c from baseline to Week 28.

Secondary endpoints

Glycaemic

Proportion of patients achieving HbA1c <7.0%.

Change in FPG and 2-hour PPG after a standardised liquid meal tolerance test (MTT) from

baseline to Week 28, and FPG from baseline to Week 2.

Body weight

Change in total body weight from baseline to Week 28.

Proportion of patients achieving weight loss ≥5.0%.

Blood pressure

Change in SBP from baseline to Week 28.

Sample size

A total of 209 patients per treatment group were required assuming a mean difference of 0.35% in HbA1c

change from baseline with EQW+dapagliflozin versus each monotherapy and a standard deviation of

1.1% and 90% power (based on a 2-sample t-test at a 0.05 significance level). Assuming a 5% drop-out

rate prior to Week 4 (Visit 6), the first visit where HbA1c was to be tested, 220 patients per treatment

arm (a total of approximately 660 patients) would have post-baseline measurements of HbA1c and thus

be included in the primary endpoint analysis. Assuming 40% screen failure, a total of 1100 patients were

to be screened.

Randomisation

Patients who met all study requirements based on inclusion and exclusion criteria were centrally

randomised in a 1:1:1 ratio at Visit 3 (Day 0) by the IVRS/IWRS to receive 1 of 3 treatments:

EQW+dapagliflozin, EQW+placebo or dapagliflozin+placebo. Randomisation was stratified by baseline

HbA1c (<9.0% or ≥9.0%).

Blinding (masking)

This was a double-blind study. Patients, the investigator, study site personnel, and Sponsor personnel

involved with data review and analysis were blinded starting at Visit 3/randomisation. Masking of

treatment assignment was achieved through the use of matching placebo injections and placebo tablets.

The trial was on-going at the time of the analyses on data collected from the 28 weeks of treatment. The

Sponsor and applicable representative(s) were unblinded in Extension Periods 1 and 2 i.e. from Week 28

onwards. Sites and patients were to remain blinded during Extension Periods 1 and 2, until the study

completion and final database lock.

Statistical methods

The Statistical Analysis Plan has not been amended and the current version (number 1.0) is dated 05

February 2016. No changes were made to the planned analyses.

The primary efficacy analysis assessed the benefit of the combination of EQW+dapagliflozin over the

individual components for the change in HbA1c from baseline to Week 28. Superiority of the combination

treatment group was to be concluded if and only if the comparisons of the combination treatment group

Assessment report

EMA/538242/2017 /51

against both control groups (EQW+placebo and dapagliflozin+placebo) were statistically significant.

The primary efficacy analysis set was the Intent-to-Treat (ITT) based on all randomised patients who

received at least 1 dose of study medication and had at least 1 post-baseline HbA1c assessment with

patients analysed as randomised.

The primary endpoint was analysed using a mixed model with repeated measures (MMRM) that included

treatment, region, baseline HbA1c stratum (<9.0% or ≥9.0%), week (Weeks 4, 8, 12 6, 20, 24, and 28),

and treatment-by-week interaction as fixed effects. Baseline HbA1c was included as a continuous

covariate. If a patient’s last available measurement during the 28-week assessment period was from an

unscheduled visit or Early Termination visit, it was mapped to the next closest scheduled visit and

included in the MMRM analysis. No other missing data imputation was performed. Data collected after the

initiation of the glycaemic rescue therapy or at the post-treatment follow-up visits after a premature

treatment discontinuation, were excluded from the analysis.

Data collected post rescue therapy and post discontinuation of study medication were included in two

sensitivity analyses using the same MMRM model as that for the primary efficacy analysis. For the

primary endpoint, supportive analyses were further performed using the same MMRM model as in the

primary analysis but based on the PP analysis set and the Randomised analysis set (all patients who

signed informed consent and were randomised). The Per-protocol (PP) set was a subset of the ITT set

excluding subjects with important protocol violation(s). In addition, an ANCOVA examining the last

available observation prior to receiving rescue therapy in ITT analysis set was conducted.

Two additional methods of sensitivity analyses were performed to compare the results from a MAR-based

(missing at random) analysis (MMRM) versus a MNAR–based analysis under several MNAR scenarios

(missing not at random). The first using copy reference (CR) or Placebo-Based Multiple Imputation which

assumes that values of patients from the experimental arm who discontinue treatment or initiate rescue

therapy will subsequently follow a trajectory of outcomes in the control arm. The second was a Tipping

Point Analysis to find a “tipping point” corresponding to a value of delta where the study conclusion of a

significant treatment effect would no longer hold.

For the secondary endpoints, the same MMRM model (as in the primary analysis) was used for continuous

variables and a stratified Cochran-Mantel-Haenszel (CMH) test was used for categorical variables. A

sequential gatekeeping procedure (Dmitrienko et al 2009) was applied to the primary and secondary

endpoints in order to control the family-wise type I error rate. If superiority of the combination treatment

group was established for the primary endpoint at 5% level of significance the set of 7 secondary

endpoints were to be tested sequentially based on a pre-defined hierarchical order.

In the analysis of proportion of patients with weight loss ≥5% and HbA1c <7.0% at Week 28 patients

with missing data were treated as non-responders. Additional analyses were performed were missing 28-

week data was imputed using last observation carried forward (LOCF).

The trial was on-going at the time of the analyses conducted when all patients had completed 28 weeks

of treatment. The analyses for the entire 52-week treatment period will be performed when all the

patients completed the Extension Period 1. Similar analyses for the entire 104-week treatment period will

be performed when all the patients completed the Extension Period 2.

Assessment report

EMA/538242/2017 /51

Results

Participant flow

Table 2 Patient Disposition - 28 Week Treatment Period (All Patients)

a Informed consent received. EQW=Exenatide 2 mg once weekly; Dapa=Dapagliflozin 10 mg QD; n=Number of patients. Source: Figure 11.1.1.1.

Recruitment

A total of 1375 patients enrolled in this study from 137 centres in Hungary, Poland, Romania, Slovakia,

South Africa and USA. The first subject was enrolled 04 September 2014 and the last subject last visit for

28-week Treatment Period was 26 April 2016.

Assessment report

EMA/538242/2017 /51

Conduct of the study

Changes in the conduct of the study

There were three amendments to the protocol dated 3 Oct 2014, 20 Feb 2015 and 2 Sept 2015. All

amendments concerned clarifications on study procedures and are not considered to affect the outcome

or interpretation of data.

No changes were made to the planned analyses.

Protocol deviations

The number of patients with important protocol deviations in each treatment group is summarized in

Table 3.

Table 3 Important protocol violations – leading to exclusion from per-protocol

analysis set – 28-week treatment period (Intent-to-treat analysis set)

Baseline data

In general, baseline demographic characteristics were similar across the treatment groups. Overall, most

patients randomized in this study were white (83.6%) and not Hispanic or Latino (60.4%). The majority

of patients in the ITT analysis set were from the US (56.8%); approximately a third were from Europe

(34.9%) and the remainder were from South Africa (8.3%).

Assessment report

EMA/538242/2017 /51

Table 4 Demographic characteristics (Intent-to-treat analysis set)

In general, baseline patient characteristics were similar across the treatment groups. The mean height of

patients in the study was 166.3 cm, and the mean weight was 90.9 kg. The mean BMI was 32.7 kg/m2,

and most patients were in the obese (≥30) BMI group (62.8%).

Table 5 Patient characteristics (ITT analysis set)

Assessment report

EMA/538242/2017 /51

In general, baseline disease characteristics were similar across the treatment groups. The mean baseline

HbA1c at the screening visit was 9.31%, and most patients were in the ≥9% HbA1c category (56.6%).

The mean duration of diabetes was 7.4 years. The mean baseline eGFR was 98 mL/min/1.73 m2.

Table 6 Disease characteristics (ITT analysis set)

Medical history

In general, medical and surgical history was similar across the treatment groups.

The most commonly reported medical history terms were hypertension (63.4%), dyslipidaemia (21.6%),

obesity (21.3%), and hyperlipidaemia (18.2%).

Assessment report

EMA/538242/2017 /51

The most commonly reported surgical history terms were hysterectomy (8.5%), cholecystectomy (7.5%),

female sterilization (6.9%), and appendectomy (6.1%).

Concomitant medication prior to study entry

The treatment groups were well balanced with regard to the use of prior concomitant medications. All

patients in the Safety analysis set took at least 1 prior concomitant medication.

The most frequently used prior concomitant medications were metformin (82.3%), metformin

hydrochloride (17.7%), acetylsalicylic acid (17.6%), and lisinopril (17.6%).

Concomitant medication added after study entry

The treatment groups were well balanced with regard to the use of new concomitant medications,

reported for 98 patients (42.4%) in the EQW+dapagliflozin group, 99 patients (43.0%) in the

EQW+placebo group, and 109 patients (46.8%) in the dapagliflozin+placebo group.

The most frequently used new concomitant medications were paracetamol (6.5%), insulin glargine

(4.6%), amoxicillin (4.0%), and ciprofloxacin (4.0%). Insulin glargine was the rescue medication used in

this study.

Use of rescue medication

A total of 36 (5.2%) patients used rescue medication: 9 patients (3.9%) in the EQW+dapagliflozin group,

10 patients (4.3%) in the EQW+placebo group, and 17 patients (7.3%) in the dapagliflozin+placebo

group.

The most frequently used rescue medication was insulin glargine (29 patients; 4.2%). Open-label titrated

basal insulin was the protocol-defined rescue medication.

Numbers analysed

The analysis sets and the number of patients in each analysis set are summarized in Table 7.

Assessment report

EMA/538242/2017 /51

Table 7 Analysis Sets – 28-week treatment period (Randomized analysis set)

Outcomes and estimation

Primary efficacy endpoint

Change in HbA1c from Baseline to Week 28

The combination of EQW+dapagliflozin was superior to EQW alone or dapagliflozin alone in reducing

HbA1c over 28 weeks.

Mean HbA1c decreased from baseline to Week 28 for all treatment groups (Table 8). The LS mean change

in HbA1c was -1.98% for the EQW+dapagliflozin group, -1.60% for the EQW+placebo group, and -1.39%

for the dapagliflozin+placebo group. The difference in LS mean change between the EQW+dapagliflozin

group and EQW+placebo group was -0.38% (p=0.004) and the difference in LS mean change between

the EQW+dapagliflozin group and dapagliflozin+placebo group was -0.59% (p<0.001).

Table 8 MMRM analysis of change in HbA1c from baseline to Week 28

(ITT analysis set)

Measurement: HbA1c

Unit: %

EQW + Dapa

(N=228)

EQW + Placebo

(N=227)

Dapa + Placebo

(N=230)

Summary statistics

n 193 184 196

Baseline mean (SD) 9.29 (1.058) 9.26 (1.080) 9.25 (1.019)

Week 28 mean (SD) 7.24 (1.280) 7.58 (1.295) 7.74 (1.130)

Adjusted change from baseline to Week 28a

LS mean (SE) -1.98 (0.094) -1.60 (0.095) -1.39 (0.092)

Assessment report

EMA/538242/2017 /51

Measurement: HbA1c

Unit: %

EQW + Dapa

(N=228)

EQW + Placebo

(N=227)

Dapa + Placebo

(N=230)

95% two-sided CI (-2.16, -1.79) (-1.79, -1.41) (-1.57, -1.21)

Difference from EQW + Dapa at Week 28a

LS mean (SE) -0.38 (0.129) -0.59 (0.127)

95% two-sided CI (-0.63, -0.13) (-0.84, -0.34)

P-valueb 0.003 <0.001

Baseline is defined as the last non-missing assessment prior to first dose. This analysis excludes measurements post rescue therapy and post premature discontinuation of study medication. a Adjusted least squares means (LS Means) and treatment group difference in the change from baseline values

at Week 28 are modelled using an MMRM method including treatment, region, baseline HbA1c stratum (<9.0% or ≥9.0%), week, and treatment by week interaction as fixed factors, and baseline value as a covariate.

b P-values are adjusted for multiplicity

CI Confidence interval; Dapa Dapagliflozin 10 mg QD; EQW Exenatide 2 mg once weekly; HbA1c Haemoglobin A1c; ITT Intent to treat; LS Least squares; MMRM Mixed model with repeated measures; n Number of patients with observed baseline and Week 28 values; N Number of patients in treatment group; SD Standard deviation; SE Standard error.

Sensitivity analyses of the primary endpoint

The results for each of the supportive and sensitivity analyses were consistent with those of the primary

analysis. In comparing the analyses, the difference in estimated treatment differences was overall small,

ranging from -0.34 to -0.38 for EQW+dapagliflozin vs EQW alone and from -0.51 to -0.59 for

EQW+dapagliflozin vs dapagliflozin alone.

Subgroup analyses of the primary endpoint

As additional analyses, the primary efficacy variable, change in HbA1c from baseline to Week 28, was

analysed by the following baseline subgroup categories: age, sex, region, ethnic group, baseline BMI,

baseline HbA1c, diabetes duration, race, and eGFR.

At Week 28, a potential treatment by age subgroup interaction was observed (<65 vs ≥65, p=0.008).

There were few patients in the ≥65 subgroup (21 to 29 per treatment group), which limits the

interpretability of the interaction term for this subgroup analysis.

Assessment report

EMA/538242/2017 /51

Table 9 MMRM analysis of HbA1c, observed and change from baseline to Week 28 by

baseline age (Intent-to-treat analysis set)

Dapa=Dapagliflozin 10 mg QD; EQW=Exenatide 2 mg once weekly; N=Number of patients in treatment group; n=Number of patients included in analysis; SD=Standard deviation; SE=Standard error. a Adjusted least squares means (LS Means) and treatment group difference in the change from baseline values

at Week 28 are modeled using a mixed model with repeated measures (MMRM) including treatment, region, baseline HbA1c stratum (<9.0% or ≥9.0%), week, subgroup, treatment by week, subgroup-by-week, subgroup-by-treatment, and subgroup-by-week-by-treatment interactions, as fixed factors, and baseline value as a covariate. The nominal p-value for the subgroup-by-treatment interaction is presented.

For age categories, interaction p-value is obtained for the age categories (<65 and ≥65) only. Baseline is defined as the last non-missing assessment prior to first dose. This analysis excludes measurements post rescue therapy and post study medication discontinuation.

No other potential subgroup-by-treatment interactions were observed.

In the table below is the analysis by baseline HbA1c.

Assessment report

EMA/538242/2017 /51

Table 10 MMRM Analysis of HbA1c, Observed and Change from Baseline to Week 28

by Baseline Subgroup Categories (Intent-to-treat Analysis Set)

EQW=Exenatide 2 mg once weekly; Dapa=Dapagliflozin 10 mg QD.

SD=Standard deviation; SE=Standard error; N=Number of patients in treatment group; n=Number of patients included in analysis; Adjusted least squares means (LSMeans) and treatment group difference in the change from baseline values at week 28 are modeled using a mixed model with repeated measures (MMRM) including treatment, region, baseline HbA1c stratum (<9.0% or >=9.0%), week, subgroup, treatment by week, subgroup-by-week, subgroup-by-treatment, and subgroup-by-week-by-treatment interactions, as fixed factors, and baseline value as a covariate. The nominal p-value for the subgroup-by-treatment interaction is presented. Baseline is defined as the last non-missing assessment prior to first dose. This analysis excludes measurements post rescue therapy and post study

medication discontinuation.

Secondary efficacy endpoints

Glycaemic endpoints

The EQW+dapagliflozin group had between 1.6 and 2.3 times as many patients who achieved HbA1c

<7.0% at Week 28 compared with the EQW+placebo and dapagliflozin+placebo groups, respectively

(Table 11). The combination of EQW+dapagliflozin was also superior to EQW+placebo and

dapagliflozin+placebo in reducing FPG at Weeks 2 and 28, and 2-hour PPG at Week 28 (Table 11).

Body weight endpoints

Assessment report

EMA/538242/2017 /51

The combination of EQW+dapagliflozin was superior to EQW+placebo and dapagliflozin+placebo in

reducing mean body weight and the proportion achieving weight loss ≥5% at Week 28 (Table 11). The

effect on weight loss was additive in the EQW+dapagliflozin group compared to the individual therapies.

Blood pressure

The combination of EQW+dapagliflozin was superior to EQW alone and dapagliflozin alone in reducing

systolic blood pressure over 28 weeks (Table 11). The BP-lowering effect was at least additive for

EQW+dapagliflozin compared to the individual therapies.

Table 11 Change from baseline to Week 28 in secondary efficacy endpoints

(ITT analysis set)

Parameter

Statistic

EQW + Dapa

(N=228)

EQW + Placebo

(N=227)

Dapa + Placebo

(N=230)

Body weight (kg)a

LS mean (SE) -3.55 (0.289) -1.56 (0.293) -2.22 (0.284)

LS mean differences (SE) -2.00 (0.406) -1.33 (0.400)

95% CI for LS mean differences (-2.79, -1.20) (-2.12, -0.55)

P-valueb <0.001 <0.001

FPG at Week 28 (mmol/L)a

LS mean (SE) -3.66 (0.163) -2.54 (0.168) -2.73 (0.162)



P-valueb <0.001 <0.001

2-hour PPG (mmol/L)a

LS mean (SE) -4.88 (0.226) -3.34 (0.237) -3.39 (0.228)



P-valueb <0.001 <0.001

Body weight loss (kg) ≥5.0%c

Week 28, n (%) 76 (33.3) 31 (13.7) 46 (20.0)

P-valueb <0.001 0.001

FPG at Week 2 (mmol/L) a

LS mean (SE) -2.30 (0.145) -1.17 (0.148) -1.46 (0.145)



P-valueb <0.001 <0.001

d

HbA1c <7%c

Week 28, n (%) 102 (44.7) 61 (26.9) 44 (19.1)

P-valueb <0.001 <0.001

Assessment report

EMA/538242/2017 /51

Parameter

Statistic

EQW + Dapa

(N=228)

EQW + Placebo

(N=227)

Dapa + Placebo

(N=230)

SBP (mmHg) a

LS mean (SE) -4.3 (0.80) -1.2 (0.82) -1.8 (0.79)



P-valueb 0.005 0.022

This analysis excludes measurements post rescue therapy and post of premature discontinuation study medication. Only SBP analysis includes measurements post rescue therapy and post study medication discontinuation a Adjusted LS means and treatment group difference in the change from baseline values at Week 28 are

modelled using an MMRM including treatment, region, baseline HbA1c stratum (<9.0% or ≥9.0%), week, and

treatment by week interaction as fixed factors, and baseline value as a covariate.

b Between group comparison is significant at alpha = 0.05 using the gatekeeping procedure.

c Categories are derived from continuous measurements. All patients with missing endpoint data are imputed as non-responders. Treatment comparison is based on the CMH test stratified by baseline HbA1c (<9.0% or

≥9.0%). P values are from the general association statistics

d Nominal p-value.

BP Blood pressure; CI Confidence interval; CMH Cochran-Mantel-Haenszel; Dapa Dapagliflozin 10 mg od; EQW Exenatide 2 mg once weekly; FPG Fasting plasma glucose; HbA1c Haemoglobin; ITT Intent-to-treat; LS Least squares; MMRM Mixed model with repeated measures; n Number of patients with observed baseline and Week 28 values; N Number of patients in treatment group; od Once daily; PPG Postprandial glucose; SBP Systolic blood

pressure; SD Standard deviation; SE Standard error

Exploratory efficacy endpoints

Glycaemic and weight-related exploratory endpoints were generally consistent with the results of the

primary and secondary endpoints. The p-values reported for all exploratory variables are nominal.

Changes in HbA1c over time from baseline to Week 28

The change in HbA1c over time from baseline to Week 28, is presented in (Figure 2). Mean HbA1c

continued to decrease from baseline through Weeks 12 to 16 of the study, and then remained stable for

all treatment groups through Week 28.

Figure 2 LS mean changes (SE) in HbA1c (%) over time from baseline to Week 28

(ITT analysis set)

Baseline is defined as Week 0. CFB change from baseline; Dapa Dapagliflozin 10 mg QD; EQW Exenatide 2 mg once weekly; HbA1c Haemoglobin

A1c; ITT Intent-to-treat; LS Least squares; N Number of patients; SE Standard error; vs Versus.

Change in weight over time from baseline to Week 28

The change in weight over time from baseline to Week 28, is presented graphically in Figure 3. Mean

weight continued to decrease from baseline through Week 12 of the study, and then remained relatively

stable in the groups administered the single agents, while in the combination treatment group weight

continued to decrease through Week 28.

Assessment report

EMA/538242/2017 /51

Figure 3 LS mean changes (SE) in weight (kg) over time from baseline to Week 28

(ITT analysis set)

Baseline is defined as Week 0. CFB Change from baseline; Dapa Dapagliflozin 10 mg od; EQW Exenatide 2 mg once weekly; ITT Intent-to-treat;

LS Least squares; N Number of patients; SE Standard error; vs Versus.

Proportion of patients achieving HbA1c ≤6.5% from baseline at Week 28

The proportion of patients achieving HbA1c ≤6.5% at Week 28 was 30.3% in the EQW+dapagliflozin

group, 18.5% in the EQW+placebo group, and 10.4% in the dapagliflozin+placebo group at Week 28 . A

greater proportion of patients achieved HbA1c ≤6.5% at Week 28 in the EQW+dapagliflozin group

compared to the EQW+placebo group (p=0.003), and the dapagliflozin+placebo group (p<0.001). All

patients with missing data were treated as non-responders.

Change in diastolic blood pressure from baseline at Week 28

There were no clinically relevant changes from baseline in the 3 treatment groups or statistical

differences between treatment groups for DBP change from baseline to Week 28.

Change in lipid profiles from baseline to Week 28

There were no clinically relevant changes in cholesterol parameters (total cholesterol, low-density

lipoprotein [LDL], high-density lipoprotein [HDL], or non-HDL cholesterol) in the EQW+dapagliflozin group

or in the EQW+placebo group. There were small increases in total cholesterol, LDL, and non-HDL

cholesterol in the dapagliflozin+placebo group. There was a clinically meaningful decrease in mean

triglycerides in each of the 3 treatment groups EQW+dapagliflozin group (-0.31 mmol/L) compared to the

EQW+placebo (-0.18 mmol/L) and dapagliflozin+placebo (-0.11 mmol/L) groups from baseline to

Week 28. The nominal p-values for the differences between the EQW + dapagliflozin group and the

dapagliflozin + placebo and EQW + placebo groups were 0.036 and 0.181, respectively.

Exploratory pharmacokinetic endpoint

Exenatide and dapagliflozin pharmacokinetic profiles

The exenatide and dapagliflozin pharmacokinetics (PK) in the EQW + dapagliflozin, dapagliflozin +

placebo, and EQW + placebo treatment groups were evaluated. No difference in exenatide exposure or

dapagliflozin exposure across the EQW+dapagliflozin, EQW+placebo or dapagliflozin+placebo groups were

seen. High inter-individual variability in data was observed.

Long-term efficacy

Study D5553C00003 efficacy data from the extension up to 52 weeks are summarised in the following.

In total, 81% of the 695 randomised patients completed the 52-week treatment period and 75% of

randomised patients completed treatment. The drop-out rate was highest in the EQW+placebo group

(23%) and lower in the two other groups (16% in the EQW+dapa group and 17% in the dapa+placebo

group).

Change in HbA1c from baseline to Week 52

Assessment report

EMA/538242/2017 /51

The differences between treatment groups observed at week 28 were maintained at 52 weeks.

Mean HbA1c decreased from baseline to Week 52 for all treatment groups. The LS mean change in HbA1c

was -1.75% for the EQW + dapagliflozin group, -1.38% for the EQW + placebo group, and -1.23% for

the dapagliflozin + placebo group at Week 52. The difference in LS mean change between the EQW +

dapagliflozin group and EQW + placebo group was -0.37% (p=0.006) and the difference in LS mean

change between the EQW + dapagliflozin group and dapagliflozin + placebo group was -0.52%

(p<0.001).

Figure 4 Change in HbA1c over time, LS mean (SE) – 52-Week Treatment Period

(Intent-to-treat analysis set)

Change from baseline to Week 52 in total body weight

The differences between treatment groups observed at week 28 were maintained at 52 weeks.

Figure 5 Change in weight over time, LS Mean (SE) – 52-Week Treatment Period

(Intent-to-treat analysis set)

Proportion of patients achieving HbA1c <7.0% at Week 52

Assessment report

EMA/538242/2017 /51

Larger proportions of patients achieved HbA1c <7.0% in the EQW + dapagliflozin group compared to the

EQW alone and dapagliflozin alone groups at Week 28 and Week 52.

The proportion of patients achieving HbA1c <7% at Week 52 was 33.8% in the EQW + dapagliflozin

group, 24.2% in the EQW + placebo group (p=0.024 for the treatment comparison against EQW +

dapagliflozin group), and 15.2% in the dapagliflozin + placebo group (p<0.001 for the treatment

comparison against EQW + dapagliflozin group). The difference in proportion of patients who achieved

HbA1c <7% between the EQW + dapagliflozin group and EQW + placebo group was 9.5% and between

the EQW + dapagliflozin group and dapagliflozin + placebo group was 18.6%.

All patients with missing endpoint data were treated as non-responders.

Long-term efficacy up to 104 weeks will be available 2Q 2018.

Summary of main study

The following tables summarise the efficacy results from the main studies supporting the present

application. These summaries should be read in conjunction with the discussion on clinical efficacy as well

as the benefit risk assessment (see later sections).

Table 12 Summary of Efficacy for trial D5553C00003 Title: A 28-week, Multicenter, Randomized, Double-Blind, Active-Controlled, Phase 3 Study with a 24-week Extension Phase Followed by a 52-week Extension Phase to Evaluate the Efficacy and Safety of Simultaneous Administration of Exenatide Once Weekly 2 mg and Dapagliflozin Once Daily 10 mg Compared to Exenatide Once Weekly 2 mg Alone and Dapagliflozin Once Daily 10 mg Alone in Patients with Type 2 Diabetes who have Inadequate Glycemic Control on Metformin.

Study identifier Study D5553C00003

Design Randomised, double-blind, active-controlled, multicentre, Phase 3b efficacy and safety study

Duration of main phase: 28 weeks

Duration of Run-in phase: not applicable/a 1-week placebo lead-in period

Duration of Extension phase: a 24- and subsequent 52-week extension

Hypothesis Superiority

Treatments groups

EQW+dapagliflozin

Exenatide QW 2mg/Dapagliflozin 10 mg OD Randomised 231

EQW+placebo Exenatide QW 2mg/Placebo. Randomised 231

Dapagliflozin+placebo Dapagliflozin 10 mg OD/Placebo. Randomised 233

Endpoints and definitions

Primary endpoint

HbA1c

Change in HbA1c from Baseline to Week 28

Secondary

endpoint

HbA1c<7% Patients achieving HbA1c<7% at Week 28

Secondary endpoint

Body weight

Change in body weight from Baseline to Week 28

Database lock

Results and Analysis

Analysis description

Primary Analysis

Assessment report

EMA/538242/2017 /51

Analysis population

and time point description

Intent to treat

Week 28

Descriptive statistics and estimate variability

Treatment group EQW+ dapagliflozin

EQW+placebo

Dapagliflozin+ placebo

Number of subject

228 227 230

HbA1c (LS Mean)

-1.98 -1.60 -1.39

(SE) 0.094 0.095 0.092

HbA1c<7% (n)

102 61 44

% 44.7 26.9 19.1

Body weight (LS Mean)

-3.55 -1.56 -2.22

(SE) 0.289 0.293 0.284

Effect estimate per

comparison

Primary endpoint

HbA1c

Comparison groups EQW+dapagliflozin

vs EQW+placebo

LS Mean Differences (SE) -0.38 (0.129)

95% CI (-0.63, -0.13)

P-value 0.003

Comparison groups EQW+dapagliflozin vs Dapagliflozin+placebo


95% CI (-0.84, -0.34)

P-value <0.001

Secondary endpoint

HbA1c<7%

Comparison groups EQW+dapagliflozin vs EQW+placebo

% 17.4%

P-value <0.001

Comparison groups EQW+dapagliflozin

vs Dapagliflozin+placebo

% 25.6%

P-value <0.001

Comparison groups EQW+dapagliflozin vs EQW+placebo

Secondary

endpoint Body weight


95% CI (-2.79, -1.20)

P-value <0.001

Comparison groups EQW+dapagliflozin vs Dapagliflozin+placebo


95% CI (-2.12, -0.55)

P-value <0.001

Notes

Analysis description

Assessment report

EMA/538242/2017 /51

2.3.3. Discussion on clinical efficacy

Design and conduct of clinical studies

Study D5553C00003 was a 28-week, randomised, double-blind, active-controlled, multicentre, Phase 3b

efficacy and safety study with a 24- and subsequent 52-week extension of simultaneous administration of

EQW 2 mg + dapagliflozin 10 mg od compared to EQW 2 mg + placebo and dapagliflozin 10 mg od +

placebo in patients with T2DM who had inadequate glycaemic control on metformin. The total study

duration will be 2 years; this submission presents data from the 28-week treatment period and the first

extension period up to 52 weeks.

The three-armed study designs allows for assessment of the contribution of each of the components in

the combined therapy.

The study included patients with T2DM treated with metformin only and with poor metabolic control as

reflected by a HbA1c between 8% and 12%. This is adequate considering that two glucose lowering drugs

were initiated at once. No restrictions regarding BMI were included in the criteria. The study duration is

sufficient to allow assessment of the short-term effect on HbA1c. Long-term data up to 104 weeks will be

available once the second extension period of the study have been finalised.

The rationale for choosing the combination of a GLP1-RA (exenatide) and a SGLT2-inhibitor is based on

the complementary MOAs of the two different drugs. The doses selected were based on the respective

product information. The use of metformin as the background medication is supported by current

guidelines. Criteria for starting rescue therapy (i.e. basal insulin) were in place.

The statistical considerations regarding the study design are overall acceptable. The trial was on-going at

the time of the analyses of the 28-week data; however, sites and patients were to remain blinded until

study completion and final database lock (after extension period 2). Randomisation was 1:1:1 and

stratified by baseline HbA1c (<9.0% or ≥9.0%). Efficacy data were to be continuously collected also

during rescue therapy and post-treatment follow-up. The study was seemingly sufficiently powered and

took into account the proportion of patients (assumed to be 5%) expected to be excluded from the

primary Intent-to-Treat (ITT) analysis. The sample size calculation was based on a mean difference of

0.35% in HbA1c change from baseline, EQW+dapagliflozin versus each monotherapy.

Considering planned analyses no changes were made and the Statistical Analysis plan had not been

amended. For the primary and the set of 7 secondary efficacy endpoints multiplicity was sufficiently

handled through the use of a gatekeeping testing strategy. In addition (considering two pairwise

comparisons) superiority of the combination treatment group was to be concluded if and only if the

comparisons of the combination treatment group against both control groups (EQW+placebo and

dapagliflozin+placebo) were statistically significant; hence no further multiplicity correction was needed.

To be included in the ITT set a patient had to have been randomised, should have received at least one

dose of study medication, and have at least one HbA1c post-baseline assessment. The first visit where

HbA1c was to be evaluated was at Week 4 (Visit 6). The primary endpoint and continuous secondary

endpoints were analysed using a mixed model with repeated measures (MMRM) based on the assumption

of data missing at random (MAR) i.e. that treatment response can be estimated without bias using

exclusively observed data. In the analyses, data collected after initiation of rescue therapy and after

treatment discontinuation were excluded. Several sensitivity and supportive analyses were planned and

have been presented. Among them the primary analysis repeated based on the Randomised set and

analyses using the primary MMRM model but taking data collected post rescue therapy and post

discontinuation of study medication into account. Two sensitivity analyses were performed to address

violations of the missing at random (MAR) assumption, one analysis using a Placebo-Based Multiple

Imputation and a Tipping Point Analysis.

Assessment report

EMA/538242/2017 /51

In general the study appears well conducted. Important protocol violations were relatively few (10 % of

patients), evenly distributed between treatment groups and mainly related to compliance with treatment.

In the EQW+dapagliflozin arm a total of 206/231 (89.2%) completed the 28 week treatment period. In

the EQW monotherapy arm and in the dapagliflozin monotherapy arm it was 196/231 (84.8%) and

209/233 (89.7%) respectively. Of those who discontinued the study, the most common reasons were

withdrawal by subject, lost to follow-up and adverse event.

Overall, 14.7% (102/695) discontinued study treatment; 14.3% (33/231) of patients in the

EQW+dapagliflozin group, 18.6% (43/231) of patients in the EQW+placebo group, and 11.2% (26/233)

of patients in the dapagliflozin+placebo group.

It was few patients who used rescue, more in the in the dapagliflozin+placebo group however than in the

other two arms. In the whole study population it was 36/695 (5.2%) whereof 9/231 (3.9%) in the

EQW+dapagliflozin group, 10/231 (4.3%) in the EQW+placebo group, and 17/233 (7.3%) in the

dapagliflozin+placebo group.

A similar proportion of patients in each treatment arm was excluded from the ITT analysis set; 3/231

(1.3%) in the EQW+dapagliflozin arm, 4/231 (1.7%) in the EQW+placebo arm and 3/233 (1.3%) in the

dapagliflozin+placebo arm. All patients but one in the EQW+placebo treatment group who did not receive

treatment was excluded due to missing required efficacy assessment. In the primary ITT analysis it was

however 14.3% (98/685) of the patients who did not contribute with data and overall, this implied that

15.5% (108/695) of those who had been randomised were excluded in the primary analysis.

Efficacy data and additional analyses

The primary ITT analysis included 197/228 (86.4%), 192/227 (84.6%) and 198/230 (86.1%) patients in

the EQW+dapagliflozin, EQW+placebo and dapagliflozin+placebo arm respectively.

The study included subjects in general representative of poorly controlled T2DM patients. The population

was, however, rather heterogeneous with e.g. BMI ranging from 19 to 65, HbA1c ranging from 6.6% to

12.6%. The duration of diabetes was also very variable from 2 months up to 39 years.

Inclusion was based on screening values. The MAH has clarified that only 5 of the 41 patients with

baseline HbA1c <8.0% had an HbA1c below the lower limit for inclusion at screening. With regards to

renal function, different methods to assess renal function were applied at screening compared to the

remaining part of the study. The difference between these methods explains the discrepancies observed

in all but one of the 25 cases with eGFR below <60 mL/min/1.73m2 at baseline. Thus the number of

patients actually violating inclusion criteria was low.

At baseline, all patients but two were on a metformin dose of 1500 mg or more and the maximum dose

used was 3000 mg. No patients were using basal insulin at baseline.

The difference in HbA1c mean change from baseline week 28 in the primary analysis was -0.38 (-0.63, -

0.13; p=0.003) for EQW+dapagliflozin vs EQW+placebo and -0.59 (-0.84, -0.34; p<0.001) for

EQW+dapagliflozin vs dapagliflozin+placebo.

Thus the contribution of exenatide to the effect of the combination was larger than that of dapagliflozin.

However, both monotherapies contributed with a clinically relevant effect although the dapagliflozin effect

is of somewhat borderline character.

Results of sensitivity analyses were consistent with those of the primary analysis supporting a conclusion

of superiority in favour of the combined treatment of EQW and dapagliflozin over each monotherapy. In

comparing the analyses, the difference in estimated treatment differences was overall small (the

exception being the analysis based on the PP set; not presented here), ranging from -0.34 to -0.38 for

Assessment report

EMA/538242/2017 /51

EQW+dapagliflozin vs EQW alone and from -0.51 to -0.59 for EQW+dapagliflozin vs dapagliflozin alone

implying that e.g. the exclusion of post rescue and post treatment discontinuation data had minor impact.

However, most of the sensitivity analyses were based on the same MMRM model as used in the primary

analysis and several was seemingly based on data from the same number of patients as in the primary

analysis including the analysis on all randomised patients in which estimated outcomes was identical to

the outcomes in the primary analysis. Although the number of patients excluded in the analyses was

fairly well balanced between the treatment arms, this concerned approximately 15% of randomised

patients. While both the analysis using a placebo-based multiple imputation and the Tipping Point analysis

are appreciated as such, again MMRM was used and the same proportion of patients as in the primary

analysis were excluded. None of the sensitivity analyses of the primary endpoint included all randomised

patients and it may be that none was sufficiently conservative to challenge the primary analysis. In

addition, the analyses of 4 (out of 5) continuous secondary endpoints were based on the same MMRM

model as was used in the primary analysis resulting in the exclusion of data from similar proportions of

patients as in the primary analysis. The MAH was asked to repeat analyses of the primary and continuous

secondary endpoints based on a data set with better adherence to the ITT-principle (e.g. the currently

defined (modified) ITT for which the exclusion of approximately 1.4% of randomised patients can be

accepted). The MAH clarified that analyses involved all randomised patients who received at least 1 dose

of study medication and had at least 1 on-treatment measurement and that no patient from this study

population was excluded from any of the efficacy analysis. The only new analyses provided in response to

this request were analyses of secondary endpoints including post-rescue/post study medication

discontinuation data. The results taking these data into account were consistent with the results of

analyses excluding these data. For SBP this was the primary analysis approach (and hence, the “new”

analysis provided was the same as the one already presented in e.g. the CSR). Sensitivity analyses of the

primary endpoint were requested to be performed based on all randomised patients using a conservative

imputation approach in case data was missing. In at least one analysis patients with missing endpoint

data, patients in need of rescue therapy and patients who discontinued the study and/or study treatment

should be handled as treatment failures (using baseline observation carried forward). Additional

sensitivity analyses for the primary endpoint (Change in HbA1c from baseline to Week 28) using ANCOVA

(BOCF) have now been provided. Although not based on all randomised but ITT; this is not further

pursued considering the minor difference and the statistically convincing outcomes. Irrespective of

analysis, differences between the treatments were (still) statistically convincing with outcomes appearing

to be robust; for the comparison of EQW+Dapa versus EWQ+placebo; estimates were very similar

compared to the outcome in the primary analysis. For EQW+Dapa versus Dapa+Placebo the difference

between the treatments was slightly smaller (point estimate of e.g. -0.45) compared to the outcome in

the primary analysis (point estimate -0.59). For the above sensitivity analyses it was (as requested)

clarified in how many cases endpoint data were available at week 28, and in how many cases endpoint

data were missing and imputation used.

Subgroup analyses were performed for the primary endpoint. There was a statistically significant

interaction with age. Older subjects (≥65 years) showed an overall greater effect of the combination than

younger patients (<65 years) (-2.39% vs -1.91%). The treatment effect of both monotherapies was

numerically lower in the older age group, the difference between age groups being most pronounced for

dapagliflozin. The responder analysis by age group was in line with the analysis for change from baseline

HbA1c. The lower effect of dapagliflozin in the older age group is in line with the data from previous

studies with dapagliflozin, whereas no difference in effect by age has been observed with EQW. There is

currently no apparent explanation to the higher than expected effect of the combination in the older

patients. However, due to the low number of subjects, the data has to be interpreted with caution.

The subgroup analysis by HbA1c showed that the effect of both the combination and the monotherapies

was, as expected, most pronounced in patients with HbA1c ≥9%. The outcome of the primary analysis

Assessment report

EMA/538242/2017 /51

was only preserved in patients with HbA1c 8-9%, thus the combination was superior to both

monotherapies in this group.

In the largest subgroup, i.e. patients with HbA1c ≥9%, the combination was only superior to

monotherapy with dapagliflozin.

In the subgroup with HbA1c <8%, which was the smallest subgroup, the combined treatment and

monotherapy with dapagliflozin showed a comparable effect to that observed in patients with HbA1c

8-9%, whereas a lower effect was observed in the EQW monotherapy group. Superiority of the

combination was not shown for any of the monotherapies, the number of patients with HbA1c <8 was

however low, thus the data should be interpreted with caution.

In order to better understand the impact of baseline HbA1c on the outcome, a table with data on

responders (HbA1c <7%) by baseline HbA1c (HbA1c <8%, HbA1c 8-9% and HbA1c ≥9%) was provided.

Although the change in HbA1c is most pronounced in the subgroup with HbA1c > 9% at baseline, the

responder rate in this subgroup is lower than in the subgroups with lower baseline HbA1c as may be

expected. A similar pattern, with more responders with combination therapy and less for the single

components, was observed for all three subgroups.

The combination of EQW+dapagliflozin was superior to EQW alone and dapagliflozin alone for all

secondary endpoints.

The proportion of patients achieving HbA1c <7% at Week 28 was 44.7% in the EQW+dapagliflozin group,

26.9% in the EQW+placebo group, and 19.1% in the dapagliflozin+placebo group. The difference

between the EQW+dapagliflozin group and EQW+placebo group was 17.4% (p<0.001) and between the

EQW+dapagliflozin group and dapagliflozin+placebo group was 25.6% (p<0.001). This analysis, based on

the ITT set excluding only 1.4% of randomised patients treating all patients with missing endpoint data

as non-responders, is considered to support a superior efficacy of EQW+dapagliflozin in reducing HbA1c.

Body weight decreased by -3.6 kg in the EQW+dapagliflozin group, -1.6 kg in the EQW+placebo group,

and -2.2 kg in the dapagliflozin+placebo group. Thus there was an additive effect on body weight by the

combination. This was also reflected by a significantly higher proportion of patients achieving a body

weight reduction of ≥5% in the EQW+dapagliflozin group (33%) compared to the EQW+placebo group

(14%), and the dapagliflozin+placebo group (20%).

An additive effect was also observed on SBP at Week 28. SBP decreased by -4.3 mmHg in the

EQW+dapagliflozin group, -1.2 mmHg in the EQW+placebo group, and -1.8 mmHg in the


Efficacy data up to 52 weeks has been provided. There was a slight increase in HbA1c of about 0.2% in

all treatment groups between week 28 and week 52 but an adequate effect was maintained and the

differences observed between treatment arms at week 28 were essentially maintained. The body weight

remained essentially stable in all treatment groups between week 28 and 52 with only a slight increase.

The proportion of patients achieving HbA1c < 7% decreased with about 10 % in the EQW+dapa group

and about 5% in the monotherapy groups between week 28 and 52.

2.3.4. Conclusions on the clinical efficacy

The combination of exenatide QW and dapagliflozin was superior to monotherapy with either exenatide or

dapagliflozin on top of metformin in lowering HbA1c after 28 weeks. The effect was maintained up to 52

weeks.

Assessment report

EMA/538242/2017 /51

2.4. Clinical safety

Introduction

The safety and tolerability of exenatide and dapagliflozin as both single agent therapies and in

combination with other therapies for T2DM, including metformin, were thoroughly documented and

evaluated in the clinical development programmes for BYDUREON and FORXIGA.

The most frequent adverse reactions reported with EQW were mainly gastrointestinal related (nausea

which was the most frequent reaction and associated with the initiation of treatment and decreased over

time, and diarrhoea). In addition, injection site reactions (pruritus, nodules, erythema), hypoglycaemia

(with a sulphonylurea), and headache occurred. Most adverse reactions associated with prolonged-release

exenatide were mild to moderate in intensity. The risk of hypoglycaemia is low and dependent on the

background therapy used.

Since immediate-release exenatide has been marketed, acute pancreatitis has been reported with a

frequency not known and acute renal failure has been reported uncommonly.

The most frequently reported adverse reaction with dapagliflozin treatment was hypoglycaemia, which

depended on the type of background therapy used in each study. In general the rate of minor

hypoglycaemias was low (<5%). Other adverse reactions reported are related to the glucosuric effect of

dapagliflozin such as events related to volume depletion and uro-genital infections.

Patient exposure

The mean duration of exenatide exposure was similar across the treatment groups in

Study D5553C00003. The mean duration of dapagliflozin exposure was similar across the treatment

groups.

The mean duration of exenatide/placebo exposure was 174.6 days in the EQW+dapagliflozin group,

175.0 days in the EQW+placebo group, and 183.0 days in the dapagliflozin+placebo group.

The mean duration of dapagliflozin/placebo exposure was 180.2 days in the EQW+dapagliflozin group,

179.8 days in the EQW+placebo group, and 188.5 days in the dapagliflozin+placebo group.

Adverse events

AEs and serious adverse events (SAEs) were reported by similar numbers of patients across the

treatment groups (Table 13). Adverse events related to study drug and AEs leading to discontinuation

were higher in the EQW+dapagliflozin and EQW+placebo groups than in the dapagliflozin+placebo group.

During the study, 5 subjects died.

Table 13 Overall summary of adverse events for the 28-week treatment period

(Safety analysis set)

Adverse event category Number (%) of patientsa

EQW + Dapa

(N=231)

EQW + Placebo

(N=230)

Dapa + Placebo

(N=233)

Total

(N=694)

Any AE 131 (56.7) 124 (53.9) 121 (51.9) 376 (54.2)

Any AE with outcome of death 3 (1.3) 1 (0.4) 1 (0.4) 5 (0.7)

Any SAE (including events with outcome

= death) 10 (4.3) 8 (3.5) 10 (4.3) 28 (4.0)

Any AE leading to discontinuation of

treatment 9 (3.9) 11 (4.8) 5 (2.1) 25 (3.6)

Assessment report

EMA/538242/2017 /51

Table 13 Overall summary of adverse events for the 28-week treatment period

(Safety analysis set)

Adverse event category Number (%) of patientsa

EQW + Dapa

(N=231)

EQW + Placebo

(N=230)

Dapa + Placebo

(N=233)

Total

(N=694)

Any SAE leading to discontinuation of

treatment 0 2 (0.9) 1 (0.4) 3 (0.4)

Any AE related to treatmentb 60 (26.0) 53 (23.0) 37 (15.9) 150 (21.6)

A 28-week treatment period AE is defined as an AE occurring or an existing event worsening during or after the time of

the first dose of randomised study drug through EOT. EOT refers to Week 28. For patients early discontinued, the

EOT refers to the period after the last dose + 7 days for all treatment groups.

Includes patient , who prematurely discontinued study drug prior to Week 28 and died 22 days after last dose.

All percentages are calculated based on the number of patients in the Safety analysis set within each treatment group.

Due to the ongoing nature of the study, when the database extraction was performed post unlock for 1 SAE case on date

08 July 2016 it was found that additional AEs had been added by sites 7803, 7824, 7833, 7836, and 7852. a Patients with multiple events in the same category are counted only once in that category. Patients with events in

more than 1 category are counted once in each of those categories. b Includes causally related AEs as judged by the investigator.

AE Adverse event; Dapa Dapagliflozin 10 mg QD; EOT End of the treatment; EQW Exenatide 2 mg once weekly; N

Number of patients in treatment group; SAE Serious adverse event.

Source: Module 2.7.4, Table 3

Common adverse events

The most commonly reported AEs (frequency ≥5.0% in any treatment group) by preferred term (PT)

were diarrhoea, injection site nodule, nausea, and urinary tract infection (UTI) (Table 14).

Adverse events of nausea and diarrhoea were more common in the EQW+dapagliflozin and EQW+placebo

groups compared to the dapagliflozin+placebo group. There were no relevant differences in UTI across

the 3 treatment groups.

Table 14 Most common (frequency ≥5.0%) adverse events by preferred term

–28-week treatment period (Safety analysis set)

Preferred term Number (%) of patientsa

EQW + Dapa

(N=231)

EQW + Placebo

(N=230)

Dapa + Placebo

(N=233)

Patients with any AE 131 (56.7) 124 (53.9) 121 (51.9)

Injection site nodule 18 (7.8) 14 (6.1) 12 (5.2)

Nausea 12 (5.2) 17 (7.4) 7 (3.0)

Diarrhoea 10 (4.3) 13 (5.7) 7 (3.0)

Urinary tract infection 10 (4.3) 12 (5.2) 13 (5.6)

A 28-week treatment period AE is defined as an AE occurring or an existing event worsening during or after the time of the first dose

of randomized study drug through EOT. EOT refers to Week 28. For patients early discontinued, the EOT refers to the period

after the last dose + 7 days for all treatment groups.

All percentages are calculated based on the number of patients in the analysis set within each treatment group.

MedDRA version 19.0.

Due to the ongoing nature of the study, when the database extraction was performed post unlock for 1 SAE case on date 08 July 2016

it was found that additional adverse events had been added by sites 7803, 7824, 7833, 7836, and 7852.

Most common is defined as an AE with at least 5% incidence in any treatment.

Patients with multiple events in the same category (ie, same preferred term) are counted only once in that category. Patients with

events in more than 1 category are counted once in each of those categories.

Assessment report

EMA/538242/2017 /51

a Number (%) of patients with an AE, sorted by alphabetic order for PT.

AE Adverse event; Dapa Dapagliflozin 10 mg QD; EOT End of treatment; EQW Exenatide 2 mg once weekly; N Number of

patients in treatment group; MedDRA Medical Dictionary for Regulatory Activities; PT Preferred term; SAE Serious adverse

event

Source: Module 2.7.4, Table 4

All adverse events

The SOCs with the greatest frequency of AEs (>5% in any treatment group) included: infections and

infestations (EQW+dapagliflozin, 24.2%; EQW+placebo, 26.5%; dapagliflozin+placebo, 26.2%); GI

disorders (15.6%; 15.2%; 11.6%; respectively); general disorders and administration site conditions

(13.9%; 13.5%; 9.0%;); nervous system disorders (13.0%; 5.7%; 8.2%); skin and subcutaneous tissue

disorders (5.6%; 3.0%; 3.0%); investigations (4.8%; 9.6%; 7.7%); musculoskeletal and connective

tissue disorders (4.3%; 7.4%; 7.3%); and metabolism and nutrition disorders (3.5%; 5.2%; 4.3%).

Headache, dizziness, dermatitis, and rash were more common in the EQW+dapagliflozin group (4.8%,

3.5%, 1.3%, and 0.9%, respectively) compared to the EQW+placebo (3.5%, 0.4%, 0%, and 0.4%,

respectively) and dapagliflozin+placebo groups (3.0%, 1.7%, 0%, and 0%, respectively).

One patient in the EQW+placebo group had an AE of diabetic ketoacidosis. The event was moderate in

severity and considered unrelated to study drug.

No malignancies were reported during the study other than skin neoplasms.

Adverse events reported as related to study drug

Adverse events reported as related to study drug as judged by the investigator were higher in the

EQW+dapagliflozin (26%) and EQW+placebo (23%) groups than in the dapagliflozin+placebo (15.9%)

group.

The most common AEs related to study drug were injection site nodule (EQW+dapagliflozin, 7.8%;

EQW+placebo, 5.7%; dapagliflozin+placebo, 4.7%), nausea (3.9%, 7.0%, and 1.3%, respectively),

vomiting (0.4%, 3.5%, and 1.7%, respectively), diarrhoea (1.7%, 3.0%, and 0.4%, respectively), and

injection site induration (1.7%; 2.6%; 0.9%, respectively).

Serious adverse event/deaths/other significant events

Deaths

A total of 5 patients died during the study: 3 patients (1.3%) in the EQW+dapagliflozin group (PTs:

multiple injuries, arteriosclerosis coronary artery, and toxicity to various agents), 1 patient (0.4%) in the

EQW+placebo group (myocardial infarction), and 1 patient (0.4%) in the dapagliflozin+placebo group

(ischaemic stroke). None of the deaths were considered by the investigator to be related to study drug.

Other serious adverse events

The frequency and types of SAEs during the 28-week treatment period were low and similar across

treatment groups: SAEs were reported by 10 patients (4.3%) in the EQW+dapagliflozin group, 8 patients

(3.5%) in the EQW+placebo group, and 10 patients (4.3%) in the dapagliflozin+placebo group.

More SAEs were reported in the following SOCs for the EQW+dapagliflozin group compared to the

dapagliflozin+placebo and EQW+placebo groups (>1% difference): GI disorders and hepatobiliary

disorders.

Anaphylactic reaction SAEs were reported for a total of 2 patients. The event in the EQW+placebo group

was severe in intensity, required hospitalisation, and was considered related to study drug. The event in

the dapagliflozin+placebo group was severe in intensity, required hospitalisation, and was considered not

related to study drug.

At the PT level, no SAE was reported by more than 1 patient per treatment group.

Assessment report

EMA/538242/2017 /51

Adverse events of special interest

Incidence of hypoglycaemia

Major hypoglycaemia was an event that resulted in loss of consciousness, seizure or coma (or other

mental status change consistent with neuroglycopenia), which resolved after administration of glucagon

or glucose. In addition, any event that required third party assistance to resolve because of severe

impairment in consciousness or behaviour and was associated with a plasma or capillary glucose

concentration of <3 mmol/L (54 mg/dL) was classified as major hypoglycaemia.

Minor hypoglycaemia was a non-major hypoglycaemia event that had symptoms consistent with

hypoglycaemia and had plasma or capillary glucose value of <3 mmol/L (54 mg/dL) prior to treating the

episode.

There were no events of major or minor hypoglycaemia in the study.

A total of 8 patients (3.5%) in the EQW+dapagliflozin group, 3 patients (1.3%) in the EQW+placebo

group, and 3 patients (1.3%) in the dapagliflozin+placebo group had hypoglycaemia events which were

classified as other (defined as not meeting criteria for major or minor hypoglycaemia). Glucose values for

these events ranged from 3.0 mmol/L to 4.2 mmol/L. Most events were associated with symptoms, most

events were considered mild, and there were no severe events.

Other adverse events of special interest

Adjudicated CV events

There were few confirmed adjudicated CV events in the study, with no relevant differences in event

numbers across treatment groups: a total of 1 patient in the EQW+dapagliflozin group and 2 patients

each in the EQW+placebo and dapagliflozin+placebo groups had confirmed adjudicated CV events.

Adjudicated hepatic events

During the study, 2 events met criteria for hepatic adjudication. One event was an SAE of hepatic

enzyme increased in the EQW+placebo group that met biochemical Hy’s Law criteria. The other event

was elevated alanine aminotransferase (ALT) in the dapagliflozin+placebo group. Both events were

adjudicated by a blinded hepatic adjudication committee and considered unlikely to be causally related to

study drug.

Volume depletion

There were few potentially volume depletion-related events reported in this study: 2 patients (0.9%) in

the EQW+dapagliflozin group (3 events in total, all of dehydration), no patients in the EQW+placebo

group, and 3 patients (1.3%) in the dapagliflozin+placebo group (2 events were of hypotension and one

of syncope). None of these events were SAEs or led to discontinuation, and all of these events resolved.

Pancreatitis-related events

During the study, 2 pancreatitis-related events were reported: 1 patient (0.4%) in the EQW+dapagliflozin

group, 1 patient (0.4%) in the EQW+placebo group. Neither of the events was an SAE. Both events led

to discontinuation and 1 event was considered related to study drug.

Acute renal failure-related events

Few acute renal failure-related events were reported in this study: no patients in the EQW+dapagliflozin

group, 2 patients (0.9%) in the EQW+placebo group, and 1 patient (0.4%) in the dapagliflozin+placebo

group. None were SAEs, none led to study discontinuation, none of the events were considered related to

study drug, and all events resolved.

Assessment report

EMA/538242/2017 /51

Pancreatic carcinoma related- or thyroid neoplasm-related events

No pancreatic carcinoma-related events or thyroid neoplasm-related events were reported during the

study.

Injection site-related events

There were greater numbers of injection site-related AEs in the EQW+dapagliflozin (12.1%) and

EQW+placebo (11.7%) groups than in the dapagliflozin+placebo (6.9%) group. Most of these events

were mild in intensity and considered related to study drug.

Gastrointestinal-related events

Gastrointestinal-related events were reported more commonly in the EQW+dapagliflozin (36 patients,

15.6%) and EQW+placebo (35 patients, 15.2%) groups than in the dapagliflozin+placebo (27 patients,

11.6%) group. Most GI-related events were mild or moderate in intensity. The most commonly reported

AEs were nausea and diarrhoea. Four of GI-related events were reported as SAEs in the

EQW+dapagliflozin group; all 4 events had different PTs (abdominal pain lower, anal haemorrhage,

gastrooesophageal reflux disease, and umbilical hernia).

Laboratory findings

Haematology

Patients in the EQW+dapagliflozin and dapagliflozin+placebo groups showed an increase in mean

haemoglobin and haematocrit from baseline to Week 28 (mean change from baseline to Week 28 in

haemoglobin: EQW+dapagliflozin, 3.6 g/L; and dapagliflozin+placebo, 4.6 g/L; haematocrit:

EQW+dapagliflozin 0.01, dapagliflozin+placebo 0.01). In the EQW+placebo group, changes in mean

haemoglobin and haematocrit during the 28-week treatment period were -1.5 g/L and -0.01,

respectively.

Clinical chemistry

The mean eGFR was slightly decreased in all treatment groups, and to the smallest extent in the

EQW+dapagliflozin group. At Week 28, the mean change from baseline was -0.59 mL/min/1.73 m2 in the

EQW+dapagliflozin group, -0.39 mL/min/1.73 m2 in the EQW+placebo group, and -2.04 mL/min/1.73 m2

in the dapagliflozin+placebo group.

Individual marked laboratory abnormalities

Creatinine elevations >1.5x higher than pre-treatment creatinine were reported for more patients in the

EQW+dapagliflozin group (14 patients; 6.1%) compared to the EQW+placebo (9 patients; 3.9%) and

dapagliflozin+placebo groups (9 patients; 3.9%). One patient in the EQW+dapagliflozin group had a

creatinine elevation (≥221 μmol/L). In the EQW+dapagliflozin group, all but 2 of the cases were one-

time elevations, and in 6 of 14 patients, the creatinine value remained within the normal range. One of

the creatinine elevations in the EQW+dapagliflozin group was reported as an AE and none were reported

as SAEs.

Haematocrit elevations (>0.55) were reported for 2 patients (0.9%) in the EQW+dapagliflozin group,

no patients in the EQW+placebo group, and 4 patients (1.7%) in the dapagliflozin+placebo group. None

of these events were associated with AEs.

Elevations in alanine aminotransferase or aspartate aminotransferase accompanied by elevations in total

bilirubin

One patient (E7886303) in the EQW+placebo group met biochemical Hy’s Law Criteria (ALT ≥3x upper

limit of normal (ULN) or aspartate aminotransferase (AST) ≥3x ULN and total bilirubin ≥2x ULN). The

Assessment report

EMA/538242/2017 /51

event was initially reported as an AE of elevated liver enzymes on Day 57 that was moderate in intensity

and did not require additional treatment. The investigator did not consider this event an SAE; however,

the investigator was asked to upgrade the event to an SAE based on protocol specifications. This event

was adjudicated by a blinded adjudication committee as unlikely to be causally related to study drug.

Urinalysis

Mean urine glucose/creatinine ratio increased from baseline in the EQW+dapagliflozin group, did not

increase in the EQW+placebo group, and showed the greatest increase from baseline in the

dapagliflozin+placebo group over time. These findings are consistent with the MOA of dapagliflozin.

Vital signs

Pulse rate increased from baseline in the EQW+dapagliflozin group (mean change from baseline at Week

28 was 2.1 bpm). In the EQW+placebo group, mean change from baseline was 0.7 bpm. Pulse rate

remained stable during the study in the dapagliflozin+placebo group (mean change from baseline was

-0.1 bpm).

Few patients had a documented BP value <90/55 mmHg at any time in the study (Table 15). Between

19% and 32% of patients had decreases in SBP of >20 mmHg or DBP of >10 mmHg; the incidence of

these findings was similar in the EQW+dapagliflozin and dapagliflozin+placebo groups, and slightly higher

than in the EQW+placebo group.

Table 15 Number (%) of patients with marked blood pressure abnormalities at

any time during the 28-week treatment period (Safety analysis set)

Vital signs SI

Units

Criteriona EQW + Dapa

(N=231)

EQW + Placebo

(N=230)

Dapa + Placebo

(N=233)

SBP mmHg <90 0 0 1 (0.4)

>20 Decrease 65 (28.1) 44 (19.1) 60 (25.8)

DBP mmHg <55 2 (0.9) 0 4 (1.7)

>10 Decrease 70 (30.3) 61 (26.5) 75 (32.2)

All percentages are calculated based on the number of patients within that treatment group. a Prespecified vital signs marked abnormality criteria that is considered clinically important.

Dapa Dapagliflozin 10 mg QD; DBP Diastolic blood pressure; EQW Exenatide 2 mg once weekly; SBP Systolic blood pressure. Source: Derived from CSR Table 47.

Anti-exenatide antibodies

A majority of patients in the EQW+dapagliflozin and EQW+placebo groups developed anti-exenatide

antibodies at some point over the 28-week treatment period. Patients with positive antibodies had a

higher rate of injection-site-related AEs.

Anti-exenatide antibodies were observed in 74.4% of patients in the EQW+dapagliflozin group and 76.1%

of patients in the EQW+placebo group at any time during the study. The EQW+dapagliflozin group

showed a greater percentage of patients (41.7%) with high positive antibody titres as compared to the

EQW+placebo group (28.0%). The proportion of patients positive for anti-exenatide antibodies plateaued

at Week 8 and remained stable through the end of the 28-week treatment period.

The mean change from baseline in HbA1c was similar across anti-exenatide antibody status and titre at

Week 28 in both the EQW+dapagliflozin and EQW+placebo groups. Injection site-related AEs were more

common among patients who were positive for exenatide anti-exenatide antibodies (10.8% in the

EQW+dapagliflozin group and 10.4% in the EQW+placebo group) compared to patients who were

negative (1.3% for both subgroups).

Assessment report

EMA/538242/2017 /51

Safety in special populations

No data provided.

Safety related to drug-drug interactions and other interactions

No data provided.

Discontinuation due to adverse events

The rate of discontinuations due to AEs was low in all groups. More patients discontinued due to AEs in

the EQW+dapagliflozin (3.9%) and EQW+placebo (4.8%) groups than in the dapagliflozin+placebo group

(2.1%).

The most commonly reported AE leading to discontinuation of EQW/placebo or dapagliflozin/placebo was

nausea (1.0% overall), reported by 1 patient (0.4%) in the EQW+dapagliflozin group, 5 patients (2.2%)

in the EQW+placebo group, and 1 patient (0.4%) in the dapagliflozin+placebo group. At the PT level, no

other AE leading to discontinuation was reported by more than 1 patient per treatment group.

Post marketing experience

Limited data are available from post-marketing reports of concomitant use of EQW+dapagliflozin.

2.4.1. Discussion on clinical safety

With study D5553C00003, 28 week data on the concomitant treatment with exenatide and dapagliflozin

has been provided. A total of 694 patients were included in the study out of which 231 received the

combined treatment. The mean duration of exposure to both exenatide and dapagliflozin was about 180

days with a slightly shorter exposure for active EQW than for active dapagliflozin.

The overall reporting of AEs was somewhat higher in the EQW+dapagliflozin group (57%) compared to

the EQW group (54%) and the dapagliflozin group (52%).

Treatment related AEs were more common in the EQW treated groups (EQW+dapagliflozin, 26%; EQW

23%) than in the dapagliflozin treated group (16%).

Common adverse events constituted about 40% of all AEs in the EQW+dapagliflozin group, 45% in the

EQW group and 32% in the dapagliflozin group. GI events were more common with EQW than with

dapagliflozin. Injection site nodules were observed in all treatment groups, most common with

EQW+dapagliflozin (7.8%) but common also with placebo (5.2%). UTIs were fairly balanced but most

common with dapagliflozin (5.6% vs 4.3% with EQW+dapagliflozin and 5.2% with EQW).

For the SOCs with the greatest frequency of AEs, the reporting was rather well balanced between groups.



respectively) and dapagliflozin+placebo groups (3.0%, 1.7%, 0%, and 0%, respectively). One patient in

the EQW+placebo group had an AE of diabetic ketoacidosis. No malignancies were reported during the

study other than skin neoplasms.

AEs judged as drug related were mainly GI events and injection site reactions, in line with the known

safety profile for exenatide.

Five deaths occurred during the study, 3 of which were in the EQW+dapagliflozin groups. Two of these

deaths were due to trauma/suicide. The three remaining deaths were all CV deaths, one in each

treatment group.

Assessment report

EMA/538242/2017 /51

The overall reporting of SAEs was balanced between groups. There were more SAEs in the SOC GI

disorders and hepatobiliary disorders for EQW+dapagliflozin (4 and 3) than with EQW (0 and 1) whereas

no such events were reported for dapagliflozin. Two anaphylactic reactions were reported, one in the

EQW group and one in the dapagliflozin group. At the PT level, no SAE was reported by more than 1

patient per treatment group and no specific patterns could be observed.

Hypoglycaemia was an AE of special interest in the study. As expected, the incidence of hypoglycaemia

was low and no events met the criteria of minor or major hypoglycaemia. A total of 8 patients (3.5%) in

the EQW+dapagliflozin group, 3 patients (1.3%) in the EQW+placebo group, and 3 patients (1.3%) in the

dapagliflozin+placebo group had hypoglycaemia events with glucose values ranging from 3.0 mmol/L to

4.2 mmol/L.

With regards to other AEs of special interest, CV events were few and balanced between groups. Two

patients in the EQW+dapagliflozin group and 3 patients in the dapagliflozin group reported events

potentially volume depletion-related. One pancreatitis related event was reported each in the

EQW+dapagliflozin and EQW group. Two patients in the EQW group and one in the dapagliflozin group

had events related to acute renal failure. There were no cases of pancreatic or thyroid cancer.

Comparable rates of injection site reactions were observed in the two EQW groups (12%) and 7% of

patients in the dapagliflozin+placebo group reported such events.

Gastrointestinal-related events were equally common in the two EQW groups (15.6 and 15.2%), but

rather common also in the dapagliflozin group (12%). The findings in the EQW groups were expected

based on the increased rate of GI events observed with EQW in previous studies.

Increases in haemoglobin and haematocrit were observed in both dapagliflozin groups, consistent with

changes observed in the dapagliflozin programme. Decreases in haemoglobin and haematocrit were

observed in the EQW group. A slight decrease in eGFR was observed in all three groups, most pronounced

in the dapagliflozin group. Creatinine elevations >1.5 ULN were most common in the EQW+dapagliflozin

group (6.1%) and less common in the monotherapy groups (3.9% in both groups). Many events were

one-time elevations. Haematocrit elevations (>0.55) were most common in the dapagliflozin group

(1.7%) and in the EQW+dapagliflozin group (0.9).

One patient in the EQW group met biochemical Hy´s law criteria. This was not related to treatment but to

underlying hepatobiliary disease.

An increase in heart rate was observed in the EQW treated groups with the most pronounced increase

(2.1 bpm) in the EQW+dapagliflozin group. These findings are consistent with results observed in other

EQW studies. No increase was observed in the dapagliflozin group.

Only one patient in the dapagliflozin group had a SBP <90 mmHg. Two patients in the EQW+dapagliflozin

group and 4 in the dapagliflozin group had DBP <55 at any time of the study. Between 20-30% of

patients had a >20 mmHg decrease in SBP (lowest in the EQW group) and 26-32% had a >10 mmHg

decrease in DBP (lowest rate in the EQW group).

About 75% of patients in both groups treated with EQW developed anti-exenatide antibodies at any time-

point during the study. Although a larger proportion of patients in the EQW+dapagliflozin treated group

(42%) had high positive titres at any time point compared to patients on EQW monotherapy (28%) there

was no apparent difference between treatment arms at the time-points when antibodies were measures

(week 4, 8, 12, 20 and 28). The rate of high antibody titres at study endpoint was consistent with the

rate observed in a previous comparable study with EQW.

The occurrence of antibodies did not affect the effect on HbA1c but injection site-related events were

more common in patients with ADA. Increased injection site related AEs in patients with positive

antibodies have also been observed in previous EQW studies.

Assessment report

EMA/538242/2017 /51

Safety data was also presented by age groups < 65 years and ≥ 65 years. Due to the imbalance in the

number of patients in each group (607 patients < 65 years and 87 patients ≥ 65 years) the interpretation

of data has to be made with caution. There was no apparent difference in the overall reporting of AEs

between age groups and the reporting was numerically lower in the older age. Nor was there any

apparent difference in the pattern of AEs reported. The only finding of some significance was the

observation that the decrease in eGFR observed at week 1 was more pronounced in the older age group.

The recovery was also smaller in this age group at week 28.

Apart from this finding, the data presented give no indication on a difference in the safety profile for the

combination therapy or the single components across the two age groups.

The safety data reported with the 52 week CSR did not reveal any new safety concerns. The overall

reporting of AEs was slightly higher in the EQW+dapa group and comparable in the two monotherapy

arms. The reporting of AEs was lower during the extension phase of the study and there was change in

the pattern of reported AEs. Seven SAEs were reported during the extension phase (1 in the EQW+dapa

group, 4 in the EQW+placebo group and 2 in the dapa+placebo group). Among these events, 1 event of

renal neoplasm was reported in the EQW+placebo group. There were no deaths during extension phase of

the study. There were no events of major hypoglycaemia during the extension phase. In total 24 patients

reported hypoglycaemias and the reporting was slightly higher with the combination treatment. There

was a decrease in the proportion of patients with positive antibody titers from week 28 to week 52.

2.4.2. Conclusions on clinical safety

The safety data provided with study D5553C00003 is in line with the known safety profile for exenatide

and dapagliflozin. There was no indication of added toxicity in the group on combination therapy although

the overall reporting of AEs was slightly higher in this group compared to monotherapy. No new safety

concerns that arise from the data presented (up to 52 weeks of treatment).

2.4.3. PSUR cycle

The requirements for submission of periodic safety update reports for this medicinal product are set out in

the list of Union reference dates (EURD list) provided for under Article 107c(7) of Directive 2001/83/EC

and any subsequent updates published on the European medicines web-portal.

2.5. Risk management plan

The updated RMP version 24 was initially submitted as part of this application. A consolidated updated

version of the RMP, version 27, was subsequently submitted and assessed within the procedure.

The CHMP received the following PRAC Advice on the submitted Risk Management Plan:

The PRAC considered that the risk management plan version 27 is acceptable.

The MAH is reminded that, within 30 calendar days of the receipt of the Opinion, an updated version of

Annex I of the RMP template, reflecting the final RMP agreed at the time of the Opinion should be

submitted to [email protected].

The CHMP endorsed this advice without changes.

The CHMP endorsed the Risk Management Plan version 27 with the following content:

mailto:[email protected]

Assessment report

EMA/538242/2017 /51

Safety concerns

Important identified risks

Pancreatitis

Acute renal failure

Rapid weight loss

Injection site reactions (exenatide QW)

Important potential risks

Risks associated with anti-exenatide antibodies (focus on anaphylactic-type reactions)

Cardiac events

Pancreatic cancer

Thyroid neoplasms

Administration error (exenatide QW)

Malignant neoplasm following combination treatment with insulin

Missing information Adolescents

Pregnant women

Very elderly (≥75 years old)

Patients using exenatide in combination with other agents (TZDs and insulins)

Severe gastrointestinal disease (exenatide QW)

Various degrees of impaired renal function (exenatide QW)

Hepatic impairment (exenatide QW)

Pharmacovigilance plan

Study/activity Type, title and category (1-3)

Objectives Safety concerns addressed

Status (planned, started)

Date for submission of interim or final reports (planned or actual)

H8O-MC-GWDQ/ D5551C00003

(BCB109; EXSCEL)

(CV)

Category 3

The primary objective of EXSCEL will be to evaluate the effect of exenatide QW,

used in conjunction with the current usual care for

glycaemic control, on major macrovascular events when administered to patients with T2DM

Cardiac events

Pancreatitis

Acute renal failure

Risks associated with anti-exenatide

antibodies (focus on anaphylactic-type reactions)

Pancreatic cancer

Thyroid neoplasms

Ongoing Final report (CSR)

Q4 2018

Assessment report

EMA/538242/2017 /51

Study/activity

Type, title and category (1-3)

Objectives Safety concerns

addressed

Status

(planned, started)

Date for

submission of interim or final reports (planned or

actual)

H8O-JE-EX01/D5550C00001:Byetta post- marketing

surveillance study/Prospective patient cohort

Category 3

To assess primarily the occurrence of acute pancreatitis and major adverse CV events in relation

to the exposure to exenatide BID

Pancreatitis, CV events

Ongoing Final report Q3 2020

H8O-MC-B016/ D5551N00006: An Observational

Post-Authorisation Modified Prescription-Event Monitoring Safety Study to Monitor the Safety and

Utilization of Exenatide Once Weekly (Bydureon®) in the Primary Care Setting In England

Category 3

To study the utilisation and safety of exenatide QW to treat T2DM in new user

patients (exenatide naïve) and switchers (past exenatide BID users) under normal conditions of use in primary care in England. The objective is to quantify the

incidence rate of the important identified risk of acute pancreatitis in the first 12 months after starting treatment

Pancreatitis Ongoing Interim report was conducted in

Q4 2015 with 2538 exenatide QW users

Final report when 5000

patients are available: Dependent upon enrolment

H8O-MC-B017: Incidence of Thyroid Neoplasm and Pancreatic Cancer in T2DM Patients who Initiate Bydureon®

Compared to Other Antihyperglycaemic Drugs (UK study)

Category 3

The objective of this study is to estimate and compare the incidence of thyroid neoplasm and pancreatic cancer among initiators of exenatide QW compared to other antidiabetes agents.

Primary Objectives are: (1) to estimate the absolute and relative incidence of newly diagnosed thyroid cancer among initiators of exenatide QW compared to matched

initiators of other

antidiabetes drugs – assessing events 1-year post drug initiation by duration of follow-up and drug exposure; (2) to estimate the absolute and relative incidence of

newly diagnosed pancreas cancer among initiators of exenatide QW compared to matched initiators of other antidiabetes drugs – assessing events 1-year post drug initiation by duration of

follow-up and drug exposure.

Pancreatic cancer

Thyroid neoplasms

Ongoing Risk assessment: Every two years until study ends

Interim report (when 20000

exenatide QW users are available): years (dependent upon

enrolment)

Final analysis will be performed after 55000 exenatide QW users: years

depending on enrolment

Assessment report

EMA/538242/2017 /51

Study/activity



addressed

Status

(planned, started)

Date for


actual)

BCB402/ D5551R00001: MTC Surveillance Study: A Case

Series Registry/Registry

Category 3

The objectives of this prospective active surveillance program are: (1)To establish a multicentre

registry of incident cases of MTC in adults in the US in order to characterize their medical histories and possible risk factors,

including history of treatment with EQW and other long-

acting GLP-1RAs; (2) To systematically monitor the annual incidence of MTC in the US through the NAACCR to identify any possible increase related to the

introduction of EQW and other long-acting GLP-1RAs into the US market

Medullary thyroid carcinoma

Ongoing Annual assessment report each Q1 until the

end of the study; final report : Q3 2028

H8O-MC-GWBQ (Adolescent)

(Byetta®)

Category 3

The primary objective of this study is to test the hypothesis that glycaemic control, as measured by

change in HbA1c from baseline to endpoint, with exenatide BID daily is superior (in at least 1 of the exenatide treatment arms) to that of placebo after 28 weeks of treatment in

adolescent patients with T2DM who are naïve to antidiabetes agents, or patients who are being treated with metformin, an SU, or a combination of

metformin and an SU.

Adolescents Ongoing Final report: Q2 2019

BCB114/D5551C00

002

(Adolescent)

Category 3

Primary objectives: To assess

the effect on glycaemic control, as measured by HbA1c, of exenatide QW following 14 weeks of treatment compared to

placebo in adolescents with T2DM; to evaluate the safety and tolerability of exenatide QW compared to placebo following 14 weeks of treatment in adolescents with T2DM

Adolescents Ongoing Final report:

Q1 2019

Assessment report

EMA/538242/2017 /51

Study/activity



addressed

Status

(planned, started)

Date for


actual)

H8O-MC-B015 extension (D5550R00003)

Category 3

To estimate the absolute and relative incidence of pancreatic cancer and thyroid neoplasm among exenatide

initiators relative to initiators of OADs.

Pancreatic cancer

Thyroid neoplasms

Ongoing Final Report: 2018

BID twice daily; CSR clinical study report; CV cardiovascular; EQW exenatide once weekly; GLP-1RA glucagon-like

peptide 1 receptor agonist; HbA1c haemoglobin A1c; MTC medullary thyroid carcinoma; NAACCR North American

Association of Central Cancer Registries; OAD oral antidiabetes drug; PhV pharmacovigilance; Q1 first quarter; Q3

third quarter; Q4 fourth quarter; QW once weekly; SU sulphonylurea; T2DM type 2 diabetes mellitus; UK United

Kingdom; US United States.

Risk minimisation measures

Safety concern Routine risk minimisation measures

Additional risk minimisation measures

Important identified risks

Pancreatitis Statements within Sections 4.4

(Special warnings and precautions for use) and 4.8 (Undesirable effects) of the

SmPC.

None

Acute renal failure Statements within Sections 4.4 (Special warnings and precautions for use) and 4.8

(Undesirable effects) of the SmPC.

None

Rapid weight loss Statements within Sections 4.4 (Special warnings and precautions for use) and 4.8 (Undesirable effects) of the

SmPC.

None

Injection site reactions (exenatide QW)

Product information such as product labelling and medication guide

None

Assessment report

EMA/538242/2017 /51

Safety concern Routine risk minimisation

measures

Additional risk minimisation

measures

Important potential risks

Risks associated with anti-exenatide antibodies (focus on anaphylactic-type reactions)

Statements within Sections 4.3 (Contraindications), and 4.8 (Undesirable effects) of the SmPC.

None

Cardiac events No association identified between exenatide and cardiac events to date.

None

Pancreatic cancer No association identified between exenatide and pancreatic cancer to date.

None

Thyroid neoplasms None. Section 5.3 Preclinical safety data of the SmPC describes the thyroid cancer incidence observed in rats. No

reasonable causal association between exenatide and thyroid neoplasm in humans has been identified to date.

None

Administration error (exenatide QW)

Product information such as product labelling and user

manual

None

Malignant neoplasm following

combination treatment with insulin

No association identified between exenatide and combination insulin use to

date.

None

Missing information

Adolescents Statements within Sections 4.4 (Special warnings and precautions for use) and 5.2 (Pharmacokinetic properties) of the SmPC.

None

Pregnant women Statements within Section 4.6 (Fertility, pregnancy and

lactation) of the SmPC.

None

Very elderly (≥75 years of age) Statements within Sections 5.2 (PK properties) of the SmPC.

None

Use of Exenatide in Combination

with Other Agents (TZDs and

insulins)

No differential adverse event profile has been found for patients taking exenatide in combination with other agents (TZDs and insulins).

None

Severe Gastrointestinal Disease

(exenatide QW)

Statements within Sections 4.2 (Posology and Method of

administration) and 5.2 (PK properties) of the SmPC

None

Assessment report

EMA/538242/2017 /51

Safety concern Routine risk minimisation

measures

Additional risk minimisation

measures

Various Degrees of Impaired

Renal

Function (exenatide QW)

Statements within Sections 4.2

(Posology and Method of administration), 4.4 (Special

warnings and precautions for use) and 5.2 (PK properties) of the SmPC.

None

Hepatic Impairment (exenatide

QW)

Statements within Sections 4.2

(Posology and Method of administration) and 5.2 (PK properties) of the SmPC.

None

QW once weekly; PK Pharmacokinetic; SmPC Summary of Product Characteristics; TZD thiazolidinedione.

2.6. Update of the Product information

As a consequence of this new indication, sections 4.1 and 5.1 of the SmPC have been updated. The

Package Leaflet has been updated accordingly.

4.1 Therapeutic indications

Bydureon is indicated for treatment of type 2 diabetes mellitus in combination with

Metformin

Sulphonylurea

Thiazolidinedione

Metformin and sulphonylurea

Metformin and thiazolidinedione

in adults who have not achieved adequate glycaemic control on maximally tolerated doses of these oral

therapies.

Bydureon is indicated in adults 18 years and older with type 2 diabetes mellitus to improve glycaemic

control in combination with other glucose-lowering medicinal products when the therapy in use, together

with diet and exercise, does not provide adequate glycaemic control (see section 4.4, 4.5 and 5.1 for

available data on different combinations).

For changes in the SmPC section 5.1 Pharmacodynamic properties see full PI in attachment.

Minor editorial changes were also made to the PI and accepted by the CHMP.

2.6.1. User consultation

No justification for not performing a full user consultation with target patient groups on the package

leaflet has been submitted by the MAH. However, the changes to the package leaflet are minimal and do

not require user consultation with target patient groups.

Assessment report

EMA/538242/2017 /51

3. Benefit-Risk Balance

Benefits

Beneficial effects

Study D5553C00003 was a 28-week, randomised, double-blind, active-controlled, multicentre, Phase 3b

efficacy and safety study of simultaneous administration of exenatide (EQW) 2 mg + dapagliflozin 10 mg

od compared to EQW 2 mg + placebo and dapagliflozin 10 mg od + placebo in patients with T2DM. The

patients were to have inadequate glycaemic control on metformin as reflected by HbA1c between 8% and

12%. The total study duration will be 2 years; this submission presents data from the 28-week main

treatment period. The rationale for choosing the combination of a GLP1-RA (exenatide) and a SGLT2-

inhibitor (dapagliflozin) is based on the complementary MOAs of the two different drugs. The three-armed

study designs allows for assessment of the contribution of each of the components to the effect of the

combined therapy.

The primary objective of the study was met since the difference in HbA1c mean change from baseline

week 28 in the primary analysis was -0.38 (-0.63, -0.13; p=0.003) for EQW+dapagliflozin vs

EQW+placebo and -0.59 (-0.84, -0.34; p<0.001) for EQW+dapagliflozin vs dapagliflozin+placebo. Thus

superiority of the combined treatment over each of the monotherapies was shown. The contribution of

exenatide (-0.59%) to the effect of the combination was larger than that of dapagliflozin (-0.38%).

The proportion of patients achieving HbA1c <7% at Week 28 was 44.7% in the EQW+dapagliflozin group,

26.9% in the EQW+placebo group, and 19.1% in the dapagliflozin+placebo group. The difference

between the EQW+dapagliflozin group and EQW+placebo group was 17.4% (p<0.001) and between the

EQW+dapagliflozin group and dapagliflozin+placebo group was 25.6% (p<0.001).

Subgroup analyses were performed for the primary endpoint. There was a statistically significant

interaction with age. Older patients (≥65 years) showed an overall greater effect of the combination than

younger patients (<65 years) (-2.39% vs -1.91%). The effect of both monotherapies was lower in older

patients compared to younger patients, with the largest difference between age groups observed with

dapagliflozin.

The subgroup analysis by HbA1c showed that the effect of both the combination and the monotherapies

was, as expected, most pronounced in patients with HbA1c ≥9%. The outcome of the primary analysis

was only preserved in patients with HbA1c 8-9%, thus the combination was superior to both

monotherapies in this group. In the subgroup of patients with HbA1c ≥9%, which constitutes the largest

subgroup, the combination was only superior to monotherapy with dapagliflozin and in the smallest

subgroup with HbA1c <8%, superiority of the combination was not shown over any of the monotherapies.

A similar pattern, with more responders with combination therapy and less for the single components,

was observed for all three subgroups. Although the change in HbA1c is most pronounced in the subgroup

with HbA1c > 9% at baseline, the responder rate in this subgroup is lower than in the subgroups with

lower baseline HbA1c as may be expected.

Body weight decreased with -3.6 kg in the EQW+dapagliflozin group, with -1.6 kg in the EQW+placebo

group, and with -2.2 kg in the dapagliflozin+placebo group. Thus there was an additive effect on body

weight by the combination.

An additive effect was also observed on SBP at Week 28. SBP decreased by -4.3 mmHg in the

EQW+dapagliflozin group, -1.2 mmHg in the EQW+placebo group, and -1.8 mmHg in the


Efficacy data up to 52 weeks has been provided. There was a slight increase in HbA1c of about 0.2% in

all treatment groups between week 28 and week 52 but an adequate effect was maintained and the

Assessment report

EMA/538242/2017 /51

differences observed between treatment arms at week 28 were essentially maintained. The body weight

remained essentially stable in all treatment groups between week 28 and 52 with only a slight increase.

The proportion of patients achieving HbA1c < 7% decreased with about 10 % in the EQW+dapa group

and about 5% in the monotherapy groups between week 28 and 52.

Uncertainty in the knowledge about the beneficial effects

The duration of the favourable effects has only been studied up to 52 weeks. However, at 52 weeks, only

a slight attenuation of the effect was observed.

Risks

Unfavourable effects

With study D5553C00003, 28 week data on the concomitant treatment with exenatide and dapagliflozin

has been provided. A total of 694 patients were included in the study out of which 231 received the

combined treatment. The mean duration of exposure to both exenatide and dapagliflozin was about 180

days with a slightly shorter exposure for active EQW than for active dapagliflozin.

The overall reporting of AEs was somewhat higher in the EQW+dapagliflozin group (57%) compared to

the EQW group (54%) and the dapagliflozin group (52%).

Treatment related AEs were more common in the EQW treated groups (EQW+dapagliflozin, 26%; EQW

23%) than in the dapagliflozin treated group (16%).

Common adverse events constituted about 40% of all AEs in the EQW+dapagliflozin group, 45% in the

EQW group and 32% in the dapagliflozin group. GI events were more common with EQW than with

dapagliflozin. Injection site nodules were observed in all treatment groups, most common with

EQW+dapagliflozin (7.8%) but common also with placebo (5.2%). UTIs were fairly balanced but most

common with dapagliflozin (5.6% vs 4.3% with EQW+dapagliflozin and 5.2% with EQW).

For the SOCs with the highest frequency of AEs, the reporting appeared balanced between groups.



respectively) and dapagliflozin+placebo groups (3.0%, 1.7%, 0%, and 0%, respectively). One patient in

the EQW+placebo group had an AE of diabetic ketoacidosis. No malignancies were reported during the

study other than skin neoplasms.

AEs judged as drug related were mainly GI events and injection site reactions, in line with the known

safety profile for exenatide.

The overall reporting of SAEs was balanced between groups. There were more SAEs in the SOC GI

disorders and hepatobiliary disorders for EQW+dapagliflozin (4 and 3) than with EQW (0 and 1) whereas

no such events were reported for dapagliflozin. Two anaphylactic reactions were reported, one in the

EQW group and one in the dapagliflozin group.

Hypoglycaemia was an AE of special interest in the study. As expected, the incidence of hypoglycaemia

was low and no events met the criteria of minor or major hypoglycaemia. A total of 8 patients (3.5%) in

the EQW+dapagliflozin group, 3 patients (1.3%) in the EQW+placebo group, and 3 patients (1.3%) in the

dapagliflozin+placebo group had hypoglycaemia events with glucose values ranging from 3.0 mmol/L to

4.2 mmol/L.

A comparable proportion of patients in both groups treated with EQW developed anti-exenatide antibodies

(about 75%), however, a larger proportion of patients in the EQW+dapagliflozin treated group (42%) had

Assessment report

EMA/538242/2017 /51

high positive titres compared to patients on EQW monotherapy (28%). The MAH is asked to discuss if

there is any biological rationale for this observed difference.

The occurrence of antibodies did not affect the effect on HbA1c but injection site-related events were

more common in patients with ADA.

Safety data was also presented by age groups < 65 years and ≥ 65 years. Due to the imbalance in the

number of patients in each group (607 patients < 65 years and 87 patients ≥ 65 years) the interpretation

of data has to be made with caution. There was no apparent difference in the overall reporting of AEs

between age groups and the reporting was numerically lower in the older age. Nor was there any

apparent difference in the pattern of AEs reported. The only finding of some significance was the

observation that the decrease in eGFR observed at week 1 was more pronounced in the older age group.

The recovery was also smaller in this age group at week 28. Apart from this finding, the data presented

give no indication on a difference in the safety profile for the combination therapy or the single

components across the two age groups.

The safety data reported for the extension of the study, with 52 week of exposure to the combination

treatment, did not reveal any new safety concerns.

Uncertainty in the knowledge about the unfavourable effects

The safety data available only covers 52 weeks of exposure, however, the reporting of AEs decreased

over time and no new safety concerns arose during the extension of the study.

Effects Table

Table 1. Effects Table for exenatide for treatment of adults with T2DM

Effect Description Unit EQW+ dapa

EQW+ placebo

Dapa+ placebo

Un-certainties/ Strength of evidence

Refe-rences

Favourable Effects

HbA1c Mean change in HbA1c from baseline

% -1.98 -1.60 -1.39

Treatment difference (95% CI)

-0.38 (-0.63, -0.13)

-0.59 (-0.84, -0.34)

Body

weight

Mean

change in

body weight from baseline

kg -3.55 -1.56 -2.22

Treatment difference

(95% CI)

-2.00 (-2.79, -1.20)

-1.33 (-2.12, -0.55)

Fasting plasma glucose

Mean change in FPG from baseline

mmol/L

-3.66 -2.54 -2.73

Treatment difference

(95% CI)

-1.12 (-1.55, -0.68)

-0.92 (-1.36, -0.49)

Assessment report

EMA/538242/2017 /51

Effect Description Unit EQW+

dapa

EQW+

placebo

Dapa+

placebo

Un-

certainties/ Strength of evidence

Refe-

rences

SBP

Change from baseline

mmHg -4.3 -1.2

-1.8

Treatment difference (95% CI)

-3.0 (-5.2, -0.9)

-2.4 (-4.5, -0.4)

Unfavourable Effects

Diarrhoea Incidence of Diarrhoea

% 4.3 5.7 3.0

Nausea Incidence of Nausea

% 5.2 7.4 3.0

Injection site nodule

Incidence of Injection site nodule

% 7.8 6.1 5.2

UTI Incidence of

UTI

% 4.3 5.2 5.6

ADA Incidence of Positive ADA

% 75 76 -

Incidence of

High titers

% 42 28 -

Benefit-Risk Balance

Importance of favourable and unfavourable effects

Study D5553C00003 provides short-term (28 weeks) data on combined treatment with a GLP1-RA

(exenatide) and a SGLT2-inhibitor (dapagliflozin). The population of T2DM patients included were in

rather poor metabolic control with a mean HbA1c in baseline of 9.3%. The combination therapy resulted

in a clinically relevant HbA1c reduction of almost 2%. The largest contribution appears to be from

exenatide, whereas the contribution from dapagliflozin was less prominent. However, when taking the

rate of responders into account, there was a clinically relevant increase in the rate of responders both in

comparison to EQW and dapagliflozin monotherapy.

In addition to the effects on metabolic control, additive effects on both body weight and SBP were

observed. The magnitude of these effects is considered clinically relevant, especially as longer term data

(52 week) show that this effect is maintained over time.

The safety data provided does not indicate that there is any additive toxicity or change in the safety

profile when exenatide is used concomitantly with dapagliflozin. There was a somewhat higher reporting

of AEs in the EQW+dapagliflozin treated group but the pattern of reported events did not differ compared

to the known safety profile for each of the monotherapies. A somewhat higher occurrence of high titers of

anti-exenatide antibodies was observed with combined treatment which possibly explains the higher

reporting of injection site reactions in this group.

Benefit-risk balance

The benefit risk balance for the combined use of exenatide and dapagliflozin is considered positive.

Discussion on the Benefit-Risk Balance

T2DM is a progressive disease where metabolic control is often difficult to achieve. Once life style changes

are insufficient to maintain metabolic control, metformin remains the first step in pharmaceutical

Assessment report

EMA/538242/2017 /51

treatment. Upon failure on metformin, current treatment guidelines recommend individualised treatment,

combining existing treatment options based on the patient’s needs. A majority of T2DM patients is

overweight and hypertensive, thus there is a need for treatment options which at least not aggravate

these conditions.

Concomitant treatment with exenatide and dapagliflozin resulted in a significantly greater reduction of

HbA1c when compared to exenatide or dapagliflozin alone. Both agents have weight and blood pressure

reducing properties and an additive effect was observed in the study.

The beneficial effects were achieved without any substantial change in the safety profile known for the

two agents.

The combination of exenatide with dapagliflozin could therefore be an important treatment option in

patients where further weight increase or increase in blood pressure needs to be avoided and when there

is a need for improvement of the metabolic control.

Since the initial authorisation of exenatide once weekly, the wording of the indication for medicinal

products for the treatment of diabetes has evolved, and in addition more data, including from study

D5553C00003 (Duration 8 study), has been accumulated regarding the combined use of exenatide once

weekly with other products for the treatment of diabetes representing the standard of care. Therefore,

the wording of the indication in section 4.1 of the SmPC refers now in more general terms to the

combined use of exenatide once weekly with other products for the treatment of diabetes. Although the

wording of the indication is relatively broad, the combinations studied are clearly described in section 5.1

of the SmPC.

Conclusions

The overall B/R of Bydureon is positive.

4. Recommendations

Outcome

Based on the review of the submitted data, the CHMP considers the following variation acceptable and

therefore recommends the variation to the terms of the Marketing Authorisation, concerning the following

change:

Variation accepted Type Annexes

affected

C.I.6.a C.I.6.a - Change(s) to therapeutic indication(s) - Addition

of a new therapeutic indication or modification of an

approved one

Type II I and IIIB







changes in the SmPC and Package Leaflet and to update the Irish local representative information in the

Package Leaflet. Furthermore, the consolidated RMP version 27 has been agreed.

Assessment report

EMA/538242/2017 /51

The variation leads to amendments to the Summary of Product Characteristics and Package Leaflet and to

the Risk Management Plan (RMP).

5. EPAR changes

The EPAR will be updated following Commission Decision for this variation. In particular the EPAR module

8 "steps after the authorisation" will be updated as follows:

Scope







changes in the SmPC and Package Leaflet and to update the Irish local representative information in the

Package Leaflet. Furthermore, the consolidated RMP version 27 has been agreed.

Summary

Please refer to the scientific discussion Bydureon EMEA/H/C/002020/II/41.

Bydureon- INN -exenatide · FPG Fasting plasma glucose GI Gastrointestinal GLP-1 Glucagon-like...

Documents

Transcript of Bydureon- INN -exenatide · FPG Fasting plasma glucose GI Gastrointestinal GLP-1 Glucagon-like...